• 16 ounces of premium automotive grade ethylene glycol antifreeze

• 6 ounces of Redline’s “Water Wetter” (half a bottle)

and top off with

• a gallon of distilled water (about 3 quarts/liters needed)

A plastic windshield washer gallon jug works well.  You will need to replace the coolant every couple of years or you might get algae.  I’m currently using “Zerex”, but if your system is in an area with poor ventilation or in a house with children, you might consider using propylene glycol based “green” (green as in environment, not the color) antifreeze instead, it’s less toxic.

My own system is about 4 1/2 years old now, and has been used on two different motherboards/CPU sets:

• Zalman Reserator 1 (blue) with integrated Eheim 300 pump, comes with very nice blue silicone tubing, but I didn’t use the Zalman water block
• 2 Danger Den “Maze 4″ copper water blocks with Lucite tops and 3/8″ outlets, custom mounted to two Slot ‘F’ AMD Opterons with strips of brass sheet metal
• ThermalTake “Aquarius” A1983 VGA water block, nickel plated with 1/4″ side outlets, this is very thin and fits between PCI cards
• Black Ice Xtreme II radiator with 3/8″ outlets (the Zalman Reserator passive cooling isn’t enough for two CPUs and a GPU)
• 2 ball-bearing fans for the radiator (120mm), and additional chassis fans for motherboard passive components (definitely needed for a Tyan S2927)
• A 4-circuit fan speed controller.  This is an unintelligent device, merely a set of 12 volt active transistor rheostats.
• A “Digital Doc 5” fan speed controller and temperature sensor, this needs no interface to the computer (translation: works with Linux)
• A few polypropylene T’s and L’s in the tubing.
• “Arctic Silver 5” heat compound

Needless to say, my system is less than easily portable.  I consider this an “anti-theft” feature.

The two separate fan speed controllers require a bit of explanation.  I made a custom cable for the radiator fans, with two separate voltage feeds and diodes to keep the two units from parasitically vampiring power from each other.  The rheostat lets me set a minimum fan speed, which is basically enough to keep the fans running at a minimum of noise.  However the “Digital Doc 5″ unit monitors the temperature of various parts of the system, including the double radiator.  If that temperature goes over 96 degrees Fahrenheit, it kicks in it’s voltage, boosting the fan to full speed.  The system runs pretty quietly, but if it gets a little warm, the fans automatically spin up to higher speed.  I run the BOINC application, and so 4 cores of 2 Opteron 2218′s are going 100% all the time, and I’ve never had a *problem.

* News Flash! Problem!

04-JUL-2010 There was an “oh, shoot!” moment when a very small leak suddenly went gusher.  A small but noticeable amount of coolant had leaked from the bottom of the Zalman Reserator, and on investigation, the plastic hose fitting out of the bottom of the aluminum tank body snapped right off.  I believe it was clearly the source of the leak, as the inside of the fracture was stained.  The inlet also showed fracture evidence.  They lasted about 5 years?  Or was it 6?

Anyway… on the 4th of July 2010 (a big patriotic holiday here in the United States of America) and a Sunday, the Loew’s superstore is open and had brass 1/4 ” pipe to 3/8 ” hose barbs.  I discussed a sarcastic tone of sympathy with a gentleman who was helpful in the plumbing aisle, he would have rather been home playing World of Warcraft and appreciated that I needed this plumbing to fix my computer.

The coolant is wiped up, the table is wiped down, the brass fittings went into the tank threads with teflon plumber’s tape, and it’s working fine.  O’Reilly’s automotive store had more Water Wetter.  I’m slightly curious to know if I’ll see any electrochemical weirdness with the brass screwed into aluminium in glycol, but that wouldn’t seem to be as dramatic as a cleanly severed line pouring coolant onto the desk.  As I was actually placing an old cake pan under the tank when trying to figure out the source of the leak, it could have been much, much worse than it was.

I suggest replacing the plastic fittings on the bottom of Zalman Reserator cooling pumps with brass fittings if the unit is used more than a few years.

This leak began without much provocation, the unit hadn’t been moved in months.  The strain on the plastic fittings was never really that much, but it seems apparent that the constant vibration of the pump was enough to cause stress fractures over a long period of time.  I can’t blame Zalman for this, but an upscale model might replace these fittings with something a little stronger.

I salute to James’s Hand-Made Potsticker Recipe, the recipe here is pretty much my implementation of the Liu Family recipe.  It’s an excellent web page and obviously has deep wonderful family roots, something I value greatly here at Caveman Recipes Inc.  I’ve also taken a few tips from allrecipes.com who had a egg noodle dough which was typical of other sites.

Shopping list:
2 lbs ground pork *
1 large napa cabbage *
small bunch ( 6 – 8 ) green onions *
large handful of cilantro *
large eggs ( need 3 – 4 )
several large cloves of garlic **
palm-sized chunk of ginger root **
all-purpose flour ( 6 cups )
brown Chinese oyster sauce **
sesame oil **
soy
salt
black pepper
cayenne pepper

* near substitute okay
** important, do not omit

Ready, Go:

• Into the second largest mixing bowl, chop the napa cabbage, salt heavily, add some water, stir, and let sit at least 15 minutes (to let it give up it’s water).
• Chop the onions, cilantro, garlic, ginger, and add with the pork into the largest mixing bowl.
• Squeeze as much water as possible from the chopped napa with your hands (tight little fist sized cabbage balls) and add to the mix.
• Punch a crater in the center of the ‘pemmican’ and add:
  • 2 Tablespoons oyster sauce
  • 1 Tablespoon sesame oil
  • 2 teaspoons salt
  • 1 teaspoon black pepper
  • 1/4 teaspoon cayenne pepper
• Mix well and refrigerate.
• Wash the salty cabbage juice out of the second largest bowl and dry with towel.  Take a break here.  Check again to make sure you put the meat in the refrigerator.

Break. Do not drink too much coffee when working with dough.

• Sift 6 cups of flour with 1 1/2 teaspoons salt into the dry bowl (you didn’t dry it, did you, lazy bum)
• Beat three eggs with a cup of warm water and add to the flour.
• Mix, add water one teaspoon at a time until a workable dough is formed.
• Knead for 5 minutes.

Break time.

• The dough should be very stiff but workable and elastic.  You can use a pin, but it’s tough work.  Roll out the dough as thinly as possible.   Alternately use a pasta sheet machine.
• Add a spoonful (or two) of filling.  I’m using a big cookie cutter with a crimper insert, so I lay another sheet of dough over the top, but you can cut the dough and fold it over.
• Cut the dough and crimp the edges.
• Stack them on a plate and freeze.  Turn them over a few times while freezing so they don’t stick to the plate or each other.

Cooking

These are full of raw pork, so cooking is important.  They can be steamed, baked, boiled, fried, and probably a few other ways I haven’t thought of.  See James’s recipe for the best way.

Dipping Sauce

• Sesame oil
• rice vinegar
• minced ginger
• cilantro or green onion
• Soy

Still experimenting with the sauce, but these are the common ingredients.

I’m not really sure I’m going to save any money with this.  I’m spending a lot for a very nice lamp, but economically, I’m not sure the energy and light bulb savings will be worth it.  But then again, that’s not the only reason to do anything.

This lamp will replace the aging standard fixture in the ceiling of my study in a house built in 1965.  This existing fixture is made for two Edison base light bulbs, and switched at the wall near the entry door.  Bo-rrring.

The idea is to stop the 120V wiring at the switch.  There I’ll put in a small DC power supply and a circuit inside the wall box with a special wall plate for the switches mounted on the PC board.  A three wire cable (ground, 24 volts DC, and a 0 – 5 volt square wave) will connect to the overhead fixture.  The lamp will be made of 6 pentagonal printed circuit boards connected together on the edges like the lower half of a dodecahedron (12 sided solid).  Each board will have 5 Cree XLamp MX-6 high-brightness LEDs, 6500 Kelvin (cool white color), for a total of 30 LEDs.

I selected the Cree XLamp MX-6′s for a couple of reasons… they have similar or better light output at 300mA as other LEDs that use a lot more amperage.   They are touted by Cree to have a nice long life with good reliability, retaining 70% of their luminosity after high-temperature age testing, which Cree seems to be taking very seriously even if there aren’t any international standards for it yet.  And… they are generally a dollar cheaper per LED.  That’s obviously not my biggest concern, but it’s a factor.  In reality, this board should support almost any LED with a minor change to the SMD pads.  I tried to limit costs where possible, but I didn’t skimp on the quality.

Here is the schematic for the dimmer board.  It features a triangle wave generator for the PWM signal, and uses a “Dallastat” digital potentiometer for 64-step up/down dimming.  There’s an adjustable hysteresis so that “all the way down” is really 0% and “all the way up” is really 100%.  A MAX6816 debouncer chip and a ‘D’ flip-flop are used to create an on/off signal, and these are  ‘and’ed with the PWM and buffered with an op-amp for the lamp on/off control output.

I used Paul Falstad’s circuit simulator to work out the component values for the triangle wave generator.  Here is the data to ‘import’ (in extremely small text so as not to bore everyone else) if you’re interested:

The dimmer printed circuit board is 1.25 x 2.5 inches, small enough to fit inside the single-gang electrical wall switch box.  The pushbutton on/off switch and the up/down dimmer rocker switch mount on the board, but the pushbutton switch is a whole centimeter shorter, and so it rides on a short riser board held up with standoffs.  The four square ground pads at the corners will use 2 cm standoffs to mount to a specially cut wall plate.  There’s also an interface for digital control of up/down/on/off through a home automation mechanism.  Here is a gerber-generated drawing of the dimmer board:

Anybody who’s spent a few days in Michigan in February knows I’ll be popping big static sparks off this switch, so there’s plenty of ESD diodes and varistors.

The uncredited protagonist of the lamp printed circuit board is a constant current driver called the CAT4101.  It’s very easy to use, takes the higher voltage needed to drive a string of LEDs, and features a PWM/Enable input.

This sucker is gonna generate some heat.  As many of the parts as I could get are operational up to 125 degrees centigrade… however all of the fun components start to derate over 70 degrees.  The back side of the board has been cleared for room for a TO-3 transistor style heat sink, with mounting holes.  I specifically selected a heat sink with prongs on it that I could bend, as all the heat sinks pointing into the center of a polyhedron might poke each other… and I admit that I didn’t want to actually do the solid modeling or math involved in worrying about that.  The CAT4101 LED driver will need to handle about 5 watts, so it has it’s own tin-plated surface mount ‘D2PAK’ heat sink, which I really hope will be enough.  The lamp boards can be daisy-chained together by two connectors on the back of each, up to the amperage of the power supply.  The PWM signal from the dimmer board is buffered again by a receiver op-amp, and this need only be on the first board of a chain, the others can shunt the op-amp with a solder jumper.  Another solder jumper permits the board to be used ‘always on’ without the dimmer.   There’s a big honking mounting hole in the center of the board, and again, there’s lots of ESD protection.

The 6500K white LEDs generate 100 lumens at 300 milliamps, the suggested and my projected load (Cree MX-6′s are available in warmer colors, but with less light output).  This should be 3000 lumens total, which is pretty much enough for surgery.   But… the digital potentiometer and the response of the LED driver should be near flat-line linear at 400 Hz, so I don’t have to run it full blast all the time.  It will be nice to be able to turn the lights UP when I really need them (like during teach-yourself brain surgery).

February 15th, 2010: I worked all weekend and a holiday, and I’m convinced that any further soliloquizins’ will just screw me up, so I pushed the buttons.  Everything is ordered.  The Gerber files went into Sierra ProtoExpress’s “No Touch” automatic verification system on the first try.  That’s scary.  I have no idea if this is even going to work, but it a be a good circuit board dere bubba.  I doubled up the dimmer board and ordered 2, so I’ll get 4 of them for about $100.  The lamp boards were 6 for $200.  The parts were $300.

February 18th, 2010: Most of the parts arrived, and they look pretty good.  Digi-Key did a terrific job again. The power supply is really pretty, and I can adjust it down to 22 volts.  The 5 Cree MX-6′s on each board are projected to drop 3.3V at 300mA, and the CAT4101 drivers can’t take more than 25V, so between 17 and 24 volts should be okay.  From the data sheets, the 24V ESD devices will work just fine with less than ‘typical’ voltage.  The heat sinks look like they’ll do the job, putting on the surface mount D2PAK heat sinks looks daunting, I believe I may invest in an electric griddle and try my hand at pancake-style reflow soldering.

February 19th, 2010: I have ordered a Reflow Soldering Controller kit from SparkFun Electronics.  This kit will help me use an infrared toaster oven as a reflow oven.  The LEDs and the D2PAK drivers and heat sinks will just never work out right without this, and I’ll be needing it in the future for the much finer components I’ll likely need to use for the wireless home automation.  I’ll be heading over to Target for the toaster oven.  I wonder how the wife will react to me baking semi-toxic fumes.  I guess I better clean the garage while I’m waiting for the Fedex and UPS deliveries.

February 28th, 2010: The dimmer works perfectly.  All the dimensions are good, all the parts fit.  The voltage regulator isn’t even getting warm.  I can adjust the hysteresis potentiometer so that one click up from the bottom starts the PWM square wave.  One click from the top, it’s almost completely on, and then at the top it’s fully on.  The switch risers and the standoffs measure up well, the LED in the pushbutton switch works.  If I hold the dimmer one way or the other, the Dallastat digital potentiometer clicks one notch, pauses, and then starts ticking along.   When power is lost and restored, it always comes up with the lamp off, and the Dallastat remembers it’s last setting.  I didn’t do any breadboarding to test any of this, so I’m pretty impressed with it.
Here is a scope picture of the PWM output, 1 volt / division vertical (center mark is zero), 1 millisecond / division horizontal.  It puts out a very clean 5V square wave at ~330 Hz.  This is 32 clicks (of 63 top to bottom) from the top of the Dallastat, so right around a 50% duty cycle.

Here is the built dimmer circuit attached to the power supply.  The pushbutton is lit (it has it’s own white LED inside), the black thing is the up/down rocker dimmer switch.  The four stand-offs at the corners will connect to the back of the wall switch plate.

The Reflow Toaster controller is not working out very well.  To write any kind of meaningful control program, I had to buy a compiler.  It also won’t work well with the goofy boot loader that’s on the chip, so I have to get a programmer.  This is not what I expected, and very disappointing.

So instead, I’m going to try hand-baking one of the boards.  The Cree lamps are what I’m worried about the most, but the tin heat sink for the driver transistor would be very difficult to solder by hand, and I’d likely destroy the transistor in the process.  Here’s my manual stopwatch profile:

1. cook at 175-200F for 5 minutes (pre-pre-heat)
2. bump to 275F, cook for 2 minutes (pre-heat)
3. bump to 410F.  As soon as it melts, count to 10, turn off heat and crack open the door.
4. Open the door a little wider after a minute.  Cool down shouldn’t happen faster than 2 – 3 minutes, so walk away and have a smoke.

I worry too much.  It all worked out great, and I’ve cooked up one pentagon.  And the result…

It works.  It would work better if I remembered that LEDs are polarized.  I got too excited and got two of them backwards on the first pentagon.  However, I soldered the bypasses on the two and calibrated the amperage to 300mA, so the working three are fully loaded.

It’s using a lot more wattage than I suspected.  The three working LEDs are using 7 watts according to my Kill-A-Watt meter, which measures at the plug.  It’s bright, but it’s also getting hot.  I’ll have to risk it and let it cook for a while and see if it cuts out.  The driver transistor has a thermal shut-off if it gets too warm, and I suspect it will.  I may only be able to use two or three LEDs per pentagon with this design.

On the second pentagon, I discovered a design flaw.  I’ve made the driver pins too close to the ground plane, and pins other than 5V shorted through the solder mask.  I fried the driver transistor.  When investigating this on the first pentagon, I found a solder ball behind the pins, which shorted the LED return to ground putting full amperage to the LEDs, and fried one of the LEDs.  Hard lesson learned.  The first now has two working LEDs (at least I didn’t fry the driver on that one) and the second is toast.  It’ll be difficult to rework them.  I repaired all the boards with a dremel.

I made a third pentagon, and that one is working great.  With 5 LEDs, it’s drawing only 6 watts at full amperage and full on PWM, which is much closer to what I was expecting.  It’s hot, both the back heat sink and the driver surface mount heatsink are hot, but not so hot that I can’t hold them and keep my fingers on them.  This suggests that they are working within acceptable limits, and I can turn down the amperage a little and be pretty comfortable that this circuit will last a long time.  I’ll get a temperature sensor soon and find out for sure exactly how hot they’re getting.  The amperage adjustment trimmer starts at 0.20A fully clockwise, and turned up to 0.25A is plenty bright and allows some comfortable lee way, Cree MX-6 nominal amperage is 0.30A, absolute maximum is 0.35A.

Here’s the third pentagon working

Here’s a shot of my desk

March 7th, 2010: I’ve been doing some thermal testing.  The programmer for the PIC chip came, and the toaster controller works.  SparkFun also sent the wrong PIC chips I ordered, big surprise… they got something else wrong.  I’m not sure I want to pursue it.  But the controller has a working thermocouple on it, and taped to the bottom of one LED, I ran the light at full blast under a porcelain food bowl.  I ran it long enough to stabilize the temperature, and it does get hot.  At full amperage, 300mA, it goes up to 78C.  At 250mA, it goes to 71C.  This is just over the limit, at 70C most of the active and critical parts start to derate rapidly.  With the bowl removed, it stabilizes at around 57C.  This is pretty good.  I can’t truly seal the thing inside a glass fixture, but given any amount of air flow, it should be just fine at nominal 300mA amperage.

Cree does “High Temperature Output Life” testing on their stock, they cook LEDs at 85C for 1008 hours and ensure that light output is not diminished by more than 15%.  Running at less than 60C, I’m guessing that I’ll have plenty of light from these for a long time.  My goal is about 250K hours, or 25 years.

I’ve cooked up the remaining boards.  I’ve also successfully repaired my hot air pencil, and removed the parts I messed up from the first two boards.  Replacements are on the way, I just hope I can successfully reflow them again.  A tip o’ the hat to my friend John who pointed me to Silver Circuits ( http://custompcb.com/ ) who I’ll be trying next time.

It’s continuing to use more wattage than I expected.  The third board with 5 LEDs at full dwell reports 12 watts on my Kill-A-Watt.  The first pentagon was doing 7 watts with three working LEDs, so this makes sense.  I’ve ordered the next size power supply.   Sometime I can make another lamp with only 3 or 4 pentagons and use the other one. I measured this one at 6 watts before? It’s possible that this one is goofy, I’ll have to finish the last 3 pentagons and see what they do.

Wednesday, March 10th, 2010: I am a knucklehead.  I didn’t need the new power supply (which is here and very pretty) it’s only using 7-8 watts per pentagon.  I had the Kill-A-Watt set to volt-amps (VA), instead of watts.  It’s pulling 12-13 VA, and 7-8 watts.  I don’t even know what a volt-amp is.  Well, I’ve got a big honking power supply for the next project and this sucker is working per specs I was expecting, woo hoo!  Replacement parts have arrived, and the Fluke 62 Mini infrared temperature sensor gun is here, it confirms the temperatures I was seeing with the thermocouple.  Nothing in the way of final assembly now, just need the time and energy.  The weather has really warmed up, so I may even get a chance to ‘splore the attic a bit and see what’s going to have to happen there for the installation.

Monday, March 15th, 2010: Driver transistors are dropping like flies.  I’m not completely sure what I’m doing to them, but when everything else looks like it’s working perfectly but there’s just no current, I’m replacing the transistor.  I’m pretty sure it has to do with me testing the 5V supplies without the filter capacitor in place.  I’m actually getting pretty good at using the hot air rework to remove the transistor and the surface-mount heat sink, I just wish they weren’t electrically so fragile.  I’ll be adding more protection in the final version of this circuit.

I’ve wired the reflow toaster with the controller, it works.  It will need some calibration, it heats up 1 degree Celcius per second, but doesn’t really cool off very fast.  I don’t have a serial port on any of my working portables, so I’m waiting on a USB-to-serial cable so I can capture and graph the temperatures.

More spare parts, more hurry up and wait.  This is what I get for going straight to prototype without component level testing.  But hey, the dimmer worked right the first time, so it could be worse.

Thursday, March 18th, 2010: More transistors are here, and USB to serial port is here.  I plugged it into the old laptop and it recognized and started the serial port driver instantly, I love Kubuntu Karmic Koala.

I’ve also purchased an Atmel EVK1101 AT32UC30256 evaluation / demonstration circuit board, and an Atmel “Dragon” programmer/debugger.  These are for experimentation with using the One-Net protocol for wireless home automation… but this is for another posting.

So once again… barring any more stupid errors on my part, I may be able to do the final assembly this weekend.  Big maybe, baby, I’ve got the next project all lined up and ready, but there’s one other problem.  I’m thinking I can resist it long enough to get some work done.

Sunday, March 21st, 2010: It works pretty well.  I abandoned the power connectors, they were just too much trouble.  Copper foil tape also didn’t work well connecting the pentagons, I used a piece of wire in between the boards and just leaded the hell out of them.  It’s messy, but it’s pretty solid.

The lamp, all 6 pentagons assembled

This is the lamp with the LEDs at full brightness, and using a camera flash.

Here’s the whole setup: power supply, dimmer, lamp and tape measure for scale

Here’s a shot in the box with no camera flash and the lamp turned down to one click above off… 1/64th, or ~ 1.5% of maximum brightness.

Saturday, April 3rd, 2010: I had to turn down the amperage.  At nominal 0.3 amps and full brightness (100% PWM dwell), it gets too hot.  A few places were creeping above 75 degrees C, and I really have to draw the line there.  Turned down to 0.25A (individually controllable per pentagon), I haven’t seen any scanned spot go over 65 degrees C.  Turning down the amperage will cost 15% of light output, but that’s why it was over engineered in the first place.

I’ve been thinking about another design using a snap-together linear railed model rather than this polyhedral design, but again, that will need it’s own post someday, and if I ever build it.

It’s time for me to get practical on this, I have to buy a hard box to put the power supply in, and mount the sucker in the ceiling.  The weather in improving rapidly, I think I may have to get on this for this Easter weekend.

What I’ve missed so far:

• explore option to bypass the 5V regulator on daisy-chained successor lamp boards,  100mA should be enough for 6 boards or more
• move the pushbutton switch ESD components to the riser (maybe not)
• add lots more ground plane vias around the LEDs
• reverse the dimmer rocker switch!  The pushbutton switch should be above the rocker when mounted, and then the rocker should have ‘up’ make the lights brighter.  The switch is currently (electrically) upside down I must have gotten this backwards twice, because it’s working correctly.

Copyright (c) 2010, Roger Cody Venable (MDVE.NET).  Some rights reserved.

This design is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License .

Problem
I am beginning to have need of additional lighting in my house.  A few years ago, my eyes started to show age, and this is not a trend I expect to reverse itself.  It’s the 21st century, I’m bored with snapping on and off light switches as I move from room to room, so I’m building a few lamps I can use.  These will be especially helpful in stairwells, but also in my basement.

Goals

• Motion activated – this is pretty easy, if nothing in the area is moving, it probably doesn’t need to be illuminated.  This also implies adjustments for sensitivity (will it trigger for me, a cat, or a mosquito) and delay (how long does it stay on after triggering).
• Ambient light sensor – no point turning on a lamp when there’s already enough light
• Energy efficiency – uses pennies worth of juice each year.  Ideally, this works during a grid power outage, and/or is completely self sustaining through auxiliary power generation devices like solar panels, a micro-windmill, or a windbelt humdinger.  It should be bright enough to keep me from tripping over laundry, not for reading.
• Solid state – no maintenance required other than occasional dusting.  Zero recurrent costs.

Vision
Create a device which triggers one or more high-brightness white light LEDs.  Device will use low voltage DC power supply, centrally located to supply multiple devices throughout a house.  Device will be installable using a circular hole saw through existing drywall, power supplied through in-wall wiring.

Design

The schematic accommodates  a dual output (+3.3V, 3 – 4 V adjustable, max 1 amp each side) power supply chip, an ambient light sensor phototransistor, the Zilog ePIR (infrared) motion detector module, three trimmers for the module control (detector sensitivity, triggered ‘on’ time, and ambient light level), a0.1 Farad aerogel power filter capacitor, the Cree XLampXP-E LED, a normally-closed optically isolated MOSFET, an N-channel power MOSFET, a PTC (thermally active resettable fuse), a power diode, and a handful of passive support components.

This is a circuit board designed with CadSoft Eagle .  The motion detector is a Zilog ePIR module, which is quite cheap (for what it does) and very simple to use.  After a bit of reading online, I’ll be using Cree XLampXP-E (group Q5) LEDs.  There will be a fairly good power supply on board, so the circuit will need about 5.5 – 7 relatively clean DC volts at half an ampere.  Also included is an ambient light detector (the light doesn’t need to be turned on unless it’s dark), and opto-isolation between the trigger logic and the relatively high-power LED.  The LED voltage can be adjusted between 3 – 4 volts, and there is a PTC device which may help protect the LED from overload.  The Cree LED is rated for 700mA, but I’ll be running it at 400mA.

According to a very brief internet research performed by me while mildly intoxicated, a 60 watt standard Edison light bulb puts out about 900 lumens.  I’ll be using 1.3 watts to drive a 107 lumen LED at around 110%, so about 1/8th the output of conventional lighting at 1/40th the power.  Since I’m driving the LED at 4/7ths of it’s rated maximum, so it should last a long, long time:  “Based on  internal  long-term  reliability  testing, Cree projects  royal blue, blue, green and white XLamp XP-E LEDs to maintain an average of 70% lumen maintenance after 50,000 hours, provided the LED junction temperature is maintained at or below 135°C and the LED is operated with a constant current of up to 700 mA”.   Heat is a big factor, the luminosity ratings drop precipitously with heat, especially with the white ones.  For this, I have designed for a heat sink on the back of the board.

So I’m prepping to make 6 units.  I’ve ordered the parts from Digi-Key , $195.86 + shipping.  Circuit board manufacture: 7 boards from Sierra Proto Express with 10 day turn-around, $189 (provided I don’t screw it up totally and the boards can actually be used).  I can also expand this circuit to use multiple LEDs in a grid, I just have to increase the amperage and the thermal handling.  I can definitely see making 8 or 12 LED units if I like the way this works out, something that mounts on or in an existing drop-ceiling acoustic tile would be neat.  I can add in a battery charging and backup circuit, a small 6-volt motorbike battery should power a few LEDs for hours.

At ~ $70 each, I hope these live up to their ‘mean time between failure’ specifications.

Here are the gerber files I submitted for manufacture: HouseLEDs-01d.zip.  See license below.

January 13th, 2010 – sweet savior on a pogo stick but Digi-Key and the United States Postal Service are fast.  The parts are here.  I swear those 0603 capacitors are the size of a mosquito’s nose, but I have faith that I have the tools that can do it.  Or… make a big honking expensive mess trying .  The Zilog modules are cool, the LEDs were humidity packed, and the aerogel caps are scary small for their rated power.
Sierra Proto Express gave me a call, very kind of them.  Aparently they have a service where they’ll put on the more complicated parts, which makes fairly complex single-board computers a distinct possibility for me, since some of those high density quad flat packages use very very fine wires.

POV-Ray generated with Eagle 3D

Wednesday, January 27th, 2010:  Sierra Proto Express says the boards have been shipped, and they’ll be here in time for me to work this weekend, so I should know soon if this design has any merit to it.  I’ve made this board pretty much just on math, I didn’t do any testing… so I’m interested in finding out just how close to reality are my limited electrical engineering skills.

Thursday, January 28th, 2010: YAY for Fedex who got the PCBs here a day early.  I’ve built one of them and it works… except…

There is one error.  The MOSFET leaks a small but significant amount of current.  The MOSFET gate switches from 0.01 to 3.29 volts pretty cleanly when the detector is triggered, but there’s just enough leakage through the MOSFET that the LED doesn’t go completely off… so it lights up with just a slight glow even when there’s no motion detected.  It’s such a small amount that I’m not sure I’m worried about it, but I’ll measure how much current it’s leaking when I build the second one.  I’m guessing it’s only a few milliamps.

Saturday, January 30th, 2010: I made the other 5 lamps.  It was difficult, my hot air pencil is not heating properly, and two of the LEDs cracked the lens as I was heating them.  The little clear drop of plastic on these Cree LEDs is not actually solid, but kinda like a very stiff silicon gel, and when heated unevenly, it will shatter off a shard.  They still work, so I live and learn, but at $5.61 each, I was slightly disappointed.  Nobody is at fault, these are not designed for hand assembly, and I am making a modern high-efficiency silicon crystal lamp using stone knives and bear skins.

At least they’re all wired and assembled, all components are showing correct function, at least as designed.  Not the kind of work I can do a lot of, my eyes are pretty sore.  The error is more severe than I had hoped, and it doesn’t quite make sense to me.  The optoisolator pulls the MOSFET gate below the 1V minimum gate threshold voltage (I’m measuring 0.003V), and the Zero gate voltage drain current (per the data sheet) is 100 uA.  This sucka seems pretty darned bright already to be running at 0.1 milliamps, so I must be missing something.  An LED capable of generating 107 lumens at 350mA still glows fairly bright when a MOSFET leaks current of 0.1mA?  I don’t ‘get’ it. ‘On’ the MOSET voltage source/drain drop is 0.58V, ‘off’ it’s 0.09V (with a lamp voltage of 3.16V).

At any rate, here are the updated final assembly notes:

• Assemble the left side of the schematic (the power supply), including the power diode (D1), the voltage regulator LX8116 (Q1), and the voltage adjustment (R3), but skip the big power cap (C1) for now.
• Apply power.  Ensure voltage across C1 is VDC minus the forward voltage drop of the power diode (D1).  I’m using 6.00 volts, the cap voltage measured 5.35 volts, for a diode forward drop of 0.65V.
• Adjust R3 (LED voltage adjustment) fully clockwise, measure the LED voltage, rotate fully counter-clockwise and measure again.  Full clockwise is low, I got 2.74 volts; full counter-clockwise is high, I got 4.10 volts.  Leave it fully clockwise at the lowest voltage.
• Check the voltage at pin 5 of the voltage regulator to be 3.3V.
• Disconnect the power.  Attach/solder all components except for the PTC.  Watch the polarity on the light sensor (Q3) and the optoisolator (OPTOMOS1).  The LED has a tiny little + on the side that goes towards the voltage regulator.  The longer pin of Q3 goes to ground, and the marking dot on the optoisolator goes towards the center of the PCB.
• Apply power and ensure that pin 3 of the Zilog e-PIR can be adjusted from 0 to 1.6 volts with the delay trimmer (R7), at top-right.
• Apply power and ensure that pin 4 of the Zilog e-PIR can be adjusted from 0 to 1.6 volts with the sensitivity trimmer (R5), at top-center.
• Disconnect power.  Turn all 3 motion detector trimmers fully clockwise, this turns sensitivity to full, delay to several minutes, and negates the effect of the ambient light sensor.
• Attach a heat sink to the back of the PC board.  Be sure not to short or come too close to one of the through-hole pads.
• Configure an ampmeter (1A scale) across the SMD pads for the PTC. Don’t look at the LED, in fact, you might want to put a piece of tape over it during adjustments. Apply power, and employ a comedy troup of dancing clowns to trigger the motion detector.  Slowly rotate the voltage adjustment trimmer (R3) counter-clockwise until the LED is drawing 400mA (0.4A).  The power supply will go higher than 700mA, but you’ll eventually burn out the LED, so DON’T DO THAT and turn the trimmer very SLOWLY.  You can adjust this upwards to 500mA, but that’s the holding voltage for the PTC, so if you wish to use a load between 0.5A and 700mA, replace the PTC with a higher value, or a piece of wire.
• Remove the ampmeter.  Measure the LED voltage from the voltage regulator at the PTC SMD pad.  This is the maximum voltage you should use on this LED.  Make a mark on R3 so you don’t go over this voltage.
• Turn off power and install the PTC (or wire shunt).  Turn all 3 motion detector trimmers fully counter-clockwise, we don’t want the motion detector to trigger at this time.  Apply power, and the LED will still show some light.  Adjust the LED voltage trimmer (R3) clockwise until the ‘off’ state amount of light from the LED is tolerable.  The ‘on’ state will still be plenty bright.  You can also force the MOSFET off by shorting pins 3 & 4 on the outside of the optoisolator.

I found the LED to be very bright (almost painful) at 400mA, and given this load it should last a long, long time.  Using less amperage should make it run even cooler and last longer.  The heat sink I used became slightly warm pretty hot when the lamp was running for few minutes an hour, so I think the heat dissipation is adequate at this load.  All bets are off when running above 70 degrees Celcius, that’s the super critical overload evacuate all personnel meltdown imminent temperature.

This is the actual manufactured PC board, for comparison to the ‘drawing’ above.

Here is detail on the heat sink on the back, and yes, it’s supposed to be a little off-square so the voltage regulator gets a little better coverage, but none of the through-hole pads is covered.

The ’6′

Build summary
I could have probably used cheaper LEDs and just powered them through the optoisolator if I’d known the ‘off’ state was going to be a problem.  In the future, I should use a real LED driver with a ‘PWM/Enable’ pin instead of a switched MOSFET.  This was fun to make and it works, but I learned a few things making mistakes.  I’ll make another post in the future when I get to installing them in the house.

Copyright (c) 2010 MDVE.NET. Some rights reserved.
This design is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License .

Basically, you change your oil… drain the old oil, replace the filter cartridge, and put in the right amount of new oil minus one quart.  Then with the engine running, you pour in the quart of Slick 50 oil, put on the oil cap, close the hood, hop in, and drive for about 30 minutes.  This shears the Teflon and the rest of the oil together, and the Teflon gets burnished into the metal of the inside of the engine.  I’ll let it ride for a regular oil change interval (which in my case is like 6K miles), then replace the filter and oil again.

I’ve been using the stuff for 30 years.  I’ve never not been impressed.  Back when I had a pretty old car (in the 80′s I had a 1968 Ford Mustang with a 289 cid V8), the effect was just shy of miraculous, and I’ve got friends who also have seen pretty amazing results.  The internet is full of claims and counterclaims on Teflon motor oil, I’ve always only used the original Slick 50.  I also looked up the quoted magazine article by ‘Consumers Digest’, using a microfilm reader as this was before the invention of the World Wide Web.

I did the research, I gave it a try, and it’s terrific.

I do believe in the engine additive self-fulfilling prophecy… when you put some snake oil product into your car, you unconsciously alter your driving habits in tiny ways which make the claims of the product come true.  Get some macho guy pouring a pint of DDT Carburetor Ointment in your gas tank and hell… it’ll even make your girlfriend’s tits seem bigger.  I don’t use a lot of that junk, all I’ve ever seen any of it do is glop up my spark plugs, and I don’t need a pseudo-sexy sports car to get happy.  (Wait, maybe I do.  I saw a ’68 Jaguar on the road recently.)

Okay, so it seems to me that on my 30 mile 75MPH drive I didn’t have to push down the gas pedal quite as far as I previously had to push it to go the same speed.  Nothing definite, nothing scientifically provable, no… I didn’t measure the amount of travel with a micrometer… it just didn’t seem like it needed as much gas to go.  Improved gas mileage has been the most noticeable effect for me, the Mustang got 21 MPG before and more like 25 MPG after I used it (yes, back in the early 80′s I got better mileage than I do now).

I’m not selling the stuff.  I don’t care if there are tons of articles on the ‘net that say it can’t possibly work.  I’ve used it, I know it does, so there, neener neener nyeah.  I’m saving my gas receipts and writing down the mileage, so I’ll have some hard data on the gas mileage.  There won’t be any way to prove that the oil is what caused it, but *I* will know.  I’ll post the results here when I do the math.

Drawbacks?  I sometimes wonder if my engine will ever wear out so I have an excuse to buy a new car.

To the makers… Gentlemen, thank you for this wonderful product, once again, it’s paying for itself.  For me.

Here’s the schematics.  They are improving daily.  Note: all POV-Ray images on this site are just for laughs and are inherently inaccurate.

Files: THESE ARE NOT DONE!  DRAFT!  THIS DOES NOT WORK!

oscillation_overthruster.zip

“ZeroFossilFuel” has a fabulous idea in this video: http://www.youtube.com/watch?v=vKjUzsNj8NM (also see http://youtube.com/watch?v=Ru8YQ6HUwbU ).  I have not tried this yet, but the more I look at it, the more I like it.

Picture Right: Lawton style gated pulse circuit, 4″ toroid, 3/8″ ferrite rod, E-core ferrite, new meter

I was having trouble figuring out how to pickup and feed back the ‘resonance’, this is helping a lot.  The original Puharich design also addresses this.

09-AUG-2008 The preliminary circuit design is nearing completion.  I need to get breadboarding!

12-AUG-2008 I’m getting some help on the forums at WaterFuelCell.org .  Files and schematic picture updated.

17-AUG-2008 I had to split the schematic into two parts, the frequency counter is now on it’s own board.  The free version of Eagle wouldn’t handle one big board, not that I could handle it either.

25-AUG-2008 Files updated.  There are new driver transistors on the MOSFET, and the board layout is pretty close.  Currently breadboarding the Frequency Counter.  This is still DRAFT.

1-SEP-2008 Files updated.  The Frequency Counter is complete, PC boards are ordered and should be here in a week.  Now breadboarding the PLL circuit.  The Resonance Scanner is working, but sensitive.  VCO-IN is responsive to 1 volt – 4.9 volts, but can go to VCC.  The top quarter volt is VERY reactive.

15-SEP-2008 With the frequency counters built, and me ramping up my metalworking capability, I’m concentrating on the PLL now.  The best values for the PLL chip itself seem to be 0.22uF and 10K, with a 1uF lock detect cap.  I might make a ‘scan speed’ select on the resonance scanner, it’s a lot easier to adjust the scale and shift of the output when it goes fast, but I’m worried that the equipment won’t be able to handle too quick a scan speed.  I’ve got the dwell side of the 556 making a standard astable square wave, and feeding it into the PLL SIG-IN line.  When power is applied, it scans once, finds it, and locks.  It’s working well, I can ‘adjust’ the frequency to simulate skewing and it keeps track.  I’ve also worked out the driver circuit after a few different tries.  I’ve settled on a push-pull totem pole design which I’ve tested up to 100KHz with the “ST8NKy” MOSFET, this is working very well.  Schematic updated. I’d like to add a lot of goofy cool blinky lights, but I’m running out of real estate.  I’m considering a ‘hybrid’ design which uses surface mount components for lots of little indicator LEDs, they aren’t required for circuit operation but look nice.

16-SEP-2008 I’ve decided to split the safety circuits off to a daughter board, like the frequency counter.   As sad as I am to admit it, some people’s kids just aren’t going to use the safety circuits, and removing them from the main board will give me lots more room for blinky LEDs.

18-SEP-2008 Schematic updated.  PLL capacitor 0.22uF is good for 20Hz – 35KHz locking.  0.1uF is working for ~200Hz to 120KHz, but I need to do more testing on this range.  Sometimes it won’t re-lock after losing it at high frequency, but will initialize and lock if power is cycled.

21-SEP-2008 Lots and lots of blinky LEDs!  I’ve added 5 different color LEDs to the board now: red = fault, blue = pulse, green = lock, yellow = dwell, and white = gate.  I’m also adding in an LM3914N chip to drive a bar array of LEDs to the scanner voltage.  Thatt should be interesting.  It will do a sawtooth back-and-forth of the LEDs, this will allow limited adjustments to be made without an oscilloscope.  Lots of lights!  I’ve ordered 10K millicandela (so like, 10 candle) LEDs from SparkFun.com , so a big Hello to them and the China Young Sun LED Technology Co., Ltd.

24-SEP-2008 On the breadboard… I have the resonance scanner circuit working, the dwell disable signal is working, the PLL is interfaced to the scanner and dwell with logic gates, the MOSFET driver works, the pulse pickup circuit kinda works. I’ve got a store-bought dual coil choke on the pulse circuit and the MOSFET driving a resistor + LED + capacitor + the other side of the choke.
Given this primitive testing setup, I’m feeding the MOSFET output back into the pulse pickup circuit.

The ‘scanner’ display looks pretty cool, and will allow fine tuning of the scanner without a scope. That could be critical to operation in some iffy combinations.

Power on: it starts up and locks. Sometimes it scans a couple of times, other times the lock is instantaneous. Changing the value of the feedback capacitor makes it unstable (smaller caps are more stable too)… for a while… then it seems to mellow out. Freq rises slowly while doing this, which I think is some type of slow magnetization of the choke core, not significant.

I have a magnet I took out of a hard drive, pretty powerful and polarized right across the flat center on one side. I hold this magnet near the D-core of the choke and the frequency goes up . I tend to pull the choke coil out of the breadboard about this time, but if I hold it down, I can hold the magnet very close to the top… and it goes fast . When the magnet gets too close or I click it onto the coil core, it loses lock and won’t regain until I hold the choke and pull the magnet off. I’m guessing the magnet blocks the transfer of the pulse in the coil, so the lock is lost and the scanner switches into circuit. When I pull the magnet away, it scans a couple of times and locks. I’ll bet I’m demagnetizing my magnet too.

Bottom line… I can modify the environment, this changes the frequency of the pulse, and the PLL keeps lock. It’s doing what it’s supposed to do. The frequency counter is a big help with this As
far as I’m concerned, I’ve got a circuit that mostly works. I need to work out the final design of the pulse pickup amp, and it’s ready for the proto shop.

I don’t have all the LEDs hooked up (just gate and lock for now), but I’ve tested the transistor driver
for them and they’ll work. I’ll put the whole thing in a blue translucent NEMA box and it will glow .

I’ve discovered McMaster-Carr. I have good stuff on the way, delrin, nylon, stainless steel, pressure switches. I need to get outside and get to working on the test tube, I just can’t seem to get motivated to get off the breadboard now that I’m having some luck with it. I got a really nice 12VDC brass water valve from Omega.com . I’m about to send the safety circuit design off for prototyping.

24-SEP-2008 (Later) Schematic and POV-Ray updated.  I cut some tubes with the new saw, it’s… scary.

27-SEP-2008 Made test cell. Tired.

28-SEP-2008 Delrin sucks, no way to glue it, but it will be useful one day with mechanical sealing.  Remade holder out of some of the 2″ clear PVC.  It looks great!  Schematics, files, pics updated.  I added a nice trick with one of the spare 4066 gates, now the scanner display gets brighter when the PLL is not locked, and dimmed when it locks.  I ordered more from Digikey, some very nice switches and knobbies for the front panel, a bunch of forgotten resistor values, and the relays for the safety board.  When I can test the optoisolator circuits with the relay, I can send the safety board design off for prototyping.

03-OCT-2008 Files and images updated.  It’s very close.

06-OCT-2008 Okay, so the circuit is close, but everything else is still far away.  It’s not delivering the voltage through the power transformer, and I’m confused.  I’m trying distilled water in the test cell and getting almost nothing.  Yes, I actually see tiny bubbles, but with the amount of juice I’m giving it, I should be melting wires.  Very very low current, and I can’t raise it even with the dwell gate defeated.  I blew an oscilloscope probe, I think I overvolted it, but I’m not sure how.  This is confusing.  I need a proper VIC transformer, that’s for sure.  I have a small heat sink on the output MOSFET, and it doesn’t even get warm.  Maybe I already cooked it, but I don’t think so.  I have to look at this another way and go back and reread the old tomes.

I tried using the Triad power transformer as the step-up, it just doesn’t seem to be drawing much from the MOSFET.  I probably need to (minimally) heat sink the MOSFET properly, but the current draw is very small, I can’t get it to draw more than 200mA, and the regular breadboarded PLL circuit draws 100mA or so of that.  The other side of the transformer shows like 20mV (MILLI volts), maybe that transformer is just a lot screwier than I thought.

I’m using a 7.5 inch ferrite rod bifilar (two wires) choke with another few turns (third winding) as the pulse pickup, that much seems to work, but the polarity of the pulse pickup seems to be the big factor on that part of the circuit.   This choke measured like 1.75mH a side when I measured it before with the cheap meter (now with blown fuses and mostly burned out).  With the pickup coil loosely wrapped over the ferrite one way, it works great, the other way, the response is very awkward in that a signal only appears when the dwell is off.  That’s strange, I have to play with that some more, at least it seems to be working.  With the diodes across the +/- inputs to the pulse op-amp, anything bigger than about 1.5 – 1.75 volts (forward bias of the MUR800E diodes) should be clipped.  Output from that op-amp is between VCC and GND.  I can see that output pulse on the scope, it works with the dwell and a little bit more after the dwell cuts.  I’ve noticed that it’s rather stable oscillation (unless the pickup coil is reversed), for all I know, it’s working perfectly and picking up too much garbage electrical interference from computers, oscilloscopes, video monitors… the other electronics in the room.  I’ve already noticed that with no pulse coil (open leads), it readily locks onto a 60Hz line signal out of the air.

At any rate, I’m becoming convinced that the circuit is doing what it’s supposed to do, and not much will need to change even if I get the right inductor(s).  If I determine that there won’t be anything I could add to the circuit to make up for any deficiency, I may send the PLL PCB off to get made anyway.  I’ll probably add in the manual override again, but that’s about all I can think to do.

I’ll take some pictures soon, maybe that will help.

07-OCT-2008

Pictures of Oscillation Overthruster output measured with a Tektronix 475 oscilloscope.

1189.  Top trace is dwell, about 25% at 878Hz (active low) taken at TP3 in the schematic (2V,0.5ms). Bottom trace is MOSFET gate (2V,0.5ms). MOSFET is driven by 9 volt supply, dwell is from one side of a 556 dual timer running on 5 volts.

1188. Top trace is dwell, same as above.  Bottom trace is the pulse output and PLL pin 14 SIGIN (2V,0.5ms).  Notice that pulses are detected after the dwell has shut off.

1190.  Close up of one dwell cycle with PLL VCO output.  Top trace is dwell (2V,50μs).  Bottom trace is PLL VCO OUT at TP6 (2V,50μs).  Here you can see the VCO is following the pulses.  Frequency counter at the VCO OUT says about 30K, it varies from 31K cold to 29K warm.  About 33 pulses in this picture per 878Hz dwell cycle is 28974, that’s about right.

1191.  Measurement at the test cell using 1.75mH dual choke, blocking diode, but no VIC transformer (1V,50uS, GND at center)  Input DC Amperage is 90 – 120 mA, sorry my cheap meter won’t do better than that right now.  The circuit normally uses that much juice, I don’t know why it’s not pulling more power.

Obviously, there’s no bubbles.  No hydrogen here, move along.

Note: I just saw something strange.  I noticed that after I turned the power off, the test cell had some voltage still on it.  I thought the scope was just decalibrated, but on GND it was at center.  Flipped back to DC, it was still showing 1.75 volts DC offset, same as in the picture above (the line at center of the waveform).  Huh?  I disconnected one of the cell wires, and the voltage is still there.  As much as I can figure, the cell is pulling some DC from the scope, and when it gets to the forward bias of the blocking diode, it stops.  So I short the cell, and it bounces back to about 0.6 volts.  I turned the circuit on, and voltage rose slowly to around 1.8 volts DC offset.  I turned the power off again, and it stays around 1.7V.  It looks like it’s sinking, very very slowly.  It’s a little weird.  It’s actually acting like… a capacitor?  Naw, that couldn’t happen.  Okay, I disconnected the scope, shorted the cell, reconnected the scope (0 volts), and then unshorted the cell.  The voltage started rising.  It’s acting like a capacitor and charging from the oscilloscope probe leakage.  Very cool, that’s a nice clue.  If I’m right.  It’s been a long day, I might be hallucinating.

This might explain a few things, there’s not even enough impurities for the water to conduct at all, so I can’t pump any current through it.  Maybe tomorrow I’ll try tap water.  Here’s a picture of my desk, don’t make me regret this:
Left to right: roll of solder, oil can of ‘Tap Magic’ cutting fluid, one-tube-set test water fuel cell (4 inch tubes), connection box with rod inductor choke, Tektronix 475 oscilloscope, yellow cased ampmeter behind banana connectors and wires, breadboard with 7046 PLL circuit, roll of Scotch 92 kapton tape, frequency counter (30579 Hz), stereo boom microscope

16-OCT-2008 I’ve done some rewiring.  The MOSFET and regulator diode are heat sinked, the test cell is hooked up.  It’s taking amperage, not a huge amount amount, but I’m still using distilled water.  It locks too easily, I’m thinking I might add the divider chip.  For some reason, the scanner circuit is invading the locked signal and knocking it out of lock.  I’m under the impression that once locked, it should really stay that way.  It falls out of lock regularly with the scanner, usually at the top when the frequency is highest.  I need it to stop scanning when locked, maybe I can tie the lock signal into the reset on that side of the 556.

26-OCT-2008 Okay, that’s doing something.  Now the scanner stops when the PLL locks, and this has a big effect.  The Triad still isn’t putting out the voltage that it should, but I’m getting a few tiny bubbles, what I might call an hour-old Alka-Seltzer.  I hooked up the cell directly (no VIC transformer) but with chokes, and I saw some hot spot frequencies.  There was one around 44KHz, but that spot was not very stable and usually jumped to 22Khz or 11KHz fairly quickly.  Sometimes the scanning would rest at 22KHz, but more often between 10 and 11 KHz.  This is interesting in that these are (reasonable) harmonics of each other and so this would tend to indicate something good.  Sometimes the circuit would jump to these points on it’s own, but more often I would cycle the power and these occured on first scan.

I’m trying the full circuit now, with reversed chokes.  If I run it with the dwell disabled, it soaks 1.82 DC Amps total circuit.  The control circuitry uses about 100 DC milliAmps.  The secondary of the Triad power transformer I’m trying as a VIC transformer is 0.92 Henries inductance according to my half burned out VC9808+ “Sinometer”.  Both sides of the choke measure 1.8 milliHenries.  The dwell at 150Hz and “ON” about %40 draws about 0.9 DC Amps (1.20 AC Amps).

When running this way there are some spikes in the +5VDC power supply, they occur on the MOSFET on times, as would be expected.  There are thick pulses during the gating times 50 millivolts high, and thin spikes at 0.1 volt.  This is ugly, it probably deprecates performance of the PLL.  I may experiment with powering the control circuitry completely separately.  This would involve cutting a fat trace on the safety board, but nothing a dremel won’t handle.

I’m also using the divider (CD4040) chip.  The chip divides the PLL Comparator input from the VCO output by 2^(the number of the pin), so Q1 = 1/2, Q2 = 1/4, Q3 = 1/8, et cetera.  This is providing a manner to adjust something… but I’m not sure what.  The frequencies change, things move around, but not really sure if it’s helping.  It will probably be a nice option to have a switch on the front of the box for it even if it’s a LOT of wires, so I’m thinking it might be permanent.  Note that Meyer’s circuit used 4017 decade counters, and so he could divide by 10, 100, or 1000.  Not sure if I want to try that.

I just tried using the JW Miller / Bourns choke instead of the 1/2 rod choke.  It seems to be working well, the circuit does everything the same but the frequency is much lower.

Copyright (c) 2008 MDVE.NET.  Some rights reserved.
This Oscillation Overthruster design is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License .

Here is a set of simplified and unfinished circuits showing different ways to connect a water fuel cell (click to enlarge)

The first drawing is the minimalist example of the MOSFET drain-source circuit (driver circuits for MOSFET gating will not be discussed here, those can be very convoluted and the one I’m using for the Oscillation Overthruster is working just fine).  In this circuit, a MOSFET gates a positive potential to ground.  Nothing else is really needed.  However, this leaves a lot of room for stray voltages from various sources, some will damage the MOSFET if left unchecked.  A more “complete” version is shown in the augmented MOSFET output circuit, which includes
* a regulator diode, which ensures that total potential is never reversed
* a clipping diode, which “shorts out” reverse currents across the VIC primary due to magnetic feedback
* a “PTC” device, this is a “resettable fuse” which protects the MOSFET from over current at the price of a small amount of resistance
* a zener diode, this “shorts out” reverse voltages across the MOSFET drain/source.  Some MOSFETs (like the STP8NK100Z) have internal zeners which perform this function.

Many would argue that some of these components would not be very beneficial, and would even hurt the purpose of the circuit, I do not dispute this.  In particular, some experimenters would not use the clipping diode as clipping the buck feedback from the VIC primary will definitely affect performance.  I’ve added the PTC myself, as the cost of these devices is far cheaper than MOSFETs, and turning the dwell up just a little too high is very easy.  It’s all perspective, if you have a warehouse full of 20-year-old junk 10GB SCSI hard drive controller PCBs with BUZ11 MOSFETs all over them, then by all means… burn out as many MOSFETs as you want.  If you’re paying $5 apiece for them, a 40 cent protection diode starts to look pretty good.  Diodes, zeners, and PTCs that handle the wattages and voltages used here are not expensive.  Some experimenters are using MOSFETs that I cannot find for under $25.

On the Oscillation Overthruster design, I’ve allowed a lot of freedom.  Some components are not required, others can be shunted with bare wire, and I’ve allowed two sizes of PTC for mounting on the PCB.  All of the heavy transistors are TO-220 types and can be socketed and (MUST be) heat sinked.

Next on the schematic, I show the quintessential Meyer VIC circuit.  The bottom choke is adjustable, either through a wiper arm in contact with the wires along the side of the coil, or through multiple taps along the choke which can be switched manually.

The other drawings on this schematic show various ways to actually connect the water fuel cell to the MOSFET.  You can use a VIC transformer, or not.  You can use chokes, or not.  The blocking diode is recommended, but is probably also optional (try and find out!).  If you use a single choke device with two windings on it, I think it’s called a “bifilar” choke (I think it’s a term Tesla used?) and you can hook that up two ways.

Okay, lots more pictures now…

	Smaller inductors Left to right: roll of 1/2″ wide Scotch 92 kapton tape, small ‘E’ core with plastic bobbin, Epcos D-core 47mH 1.3A choke (Digikey 495-2790-ND), cheap measuring scale, JW Miller / Bourns 8116-RC 50mH 2.3A choke (Digikey M8911-ND), 3/8 x 4 inch ferrite rods
	POWERLITE C-cores with scale, label from Elna Magnetics. Left to right: wrapped AMCC-100, two parts of AMCC-320 sitting on oiled anti-corrosion “gun” paper (note light rust on end), AMCC-100 set (this set is varnished with Corona Super Dope) These are very strange things.  They are made of nanocrystalline amorphous metal in the form of extremely thin ribbons of iron alloy wrapped to make the oval, then cut through the center.  By extremely thin, I mean on the order of 25 micrometers.  These are by Metglas and I got them through Elna Magnetics.
	POWERLITE C-cores – AMCC-100 closeup in sun.  Notice the edges of the thin iron ribbons are somewhat visible on the edge.  This is painted with Corona Super Dope.
	Ferrite rods in the sun, see the shiny sparklies? Not really, it didn’t come out in the picture, but they do sparkle. Two are 3/8 inch by 4 inches, the one on the right is 1/2 inch by 7 1/2 inches.
	Toroids: large yellow T400-26D (4 inch diameter) wrapped with Scotch 92 dielectric tape and a few turns of 20 AWG enameled copper wire, and Amidon ‘pulse’ toroid, pretty much the largest one they sold. The Amidon toroid is smooth and hard, makes a nice ring when struck lightly. Honestly, I’m not sure if I can ever use these, they’re just too hard to wrap.  I can make a jig I suppose, but my attempts at locating a reasonably priced toroid winding machine were futile.
	Left: various colors and gauges of copper magnet wire. Right: 3 BUZ11 N-channel MOSFETs in the middle of a power supply PC board from a 10GB 50-pin SCSI hard disk drive, circa 1994.  These will be harvested and reused.  You thought I was joking, didn’t you?  I just have this one, not a warehouse full
	TRIAD VPS10-8000 power transformer, front and back. This is a 115VAC/230VAC to 5VAC/10VAC 80 watt power transformer.  The ‘primary’ is made to take wall voltage 115/230 volts, and step that down to 5 or 10 volts.  As shown, both sides are hooked in series, so it’s wired to take 230VAC and put out 10VAC.  I use it backwards, so my input is 12V pulsing and output is… well, messy… but usually a few hundred volts with some very high spikes.  Pulses go off the scope at 50 volts/division with 8 divisions. Off topic: get a load of the cat hair stuck in the solder flux on the bottom right tab on the back.
	Ferrite rods, one raw, one wrapped with dielectric tape and then 2 windings (and again covered with dielectric tape) plus one small ‘pulse pickup’ winding with a relatively large gauge I’m suspicious that this is too large, it measures 1.75 milliHenries a side.  When I wrapped one of the 3/8″ rods, it measured 130μH a side, which seems more likely to be closer to a good value.

Not pictured: two sheets of 12″x12″ teflon, one 1/8″ thick and one 1/16″ thick.  These are for use as insulators on the ferrites or the C-cores.  They are white and square and boring, so I didn’t take a photo of them.

So I’m trying to get the TRIAD transformer to work. It’s not cooperating. I was hoping to use relatively simple off-the-shelf parts to get this working, that’s why I’m considering it at all. I don’t think this stuff is so hard that you can ONLY do it the SAME WAY, but I might be mistaken. Lots to try before giving up on that.

When I hook it up, the voltage just goes to naught. With the circuit off, I can verify very low resistance conductivity through the coils, I can verify the diode conducts only one way, everything seems to be just fine… but the actual voltage present on the water capacitor is a few millivolts. I’ll hook this up again and take another picture.

Quick edit: the rewiring is working, not sure what to make of it yet.  Basically I’ve temporarily eliminated the hookup box I’ve been using so I can just plug wires into the breadboard.

Not everyone is going to be interested in the safety circuit.  I have mixed feelings about that, but there isn’t a way to force the issue.  Table saw manufacturers can’t keep people from removing the safety guard on the saw, this isn’t any different.  I’ve removed the circuit from the main PLL board and put it on it’s own board.  This also allows a lot more real estate on the main board, but I must stress the safety… splitting water is a hazardous business.

Files: wfc_safety.zip

I’ve worked out the design and the board, I’m working through the logic to make sure that it’s viable.  It’s an ‘active live’ design, any failure causes a ‘fault’.  A fault should occur if:

* wires are broken
* the optoisolators get burned out

And the more normal faults:

* low or high water level
* low or high pressure

Low water causes a fault, and also switches a relay for a pump or valve solenoid.  Very low or high water, wow, that shouldn’t happen, especially high water.  High pressure, yeah, definitely turn it off for a while.  Low pressure: normal when the system is first starting, but what happens when there’s a blowout?  You wouldn’t want this to just keep on running with a hole in the plumbing, so low pressure causes a fault.  To get the system running, there’s a “pilot” momentary switch which defeats the low pressure fault, this is just like the pilot switch on a gas appliance, you hold it down for a minute while the system starts, then you can let it go as soon as there’s enough pressure to keep it running.

Parts list

28-SEP-2008 Added fuses and PTC.  Board will fit 90mm spaced holes.  Schematic, files and pictures updated.

01-OCT-2008 I got the relays, tested that part of the circuit, and it’s good.  The ATO fuse clips look really good, I’m happy with the way that turned out.  Files and schematic updated, this is the version I’m submitting for manufacture to Sierra Proto Express .  It’s been accepted and is “in progress”.  6 boards should be here 09-OCT-2008.

Copyright (c) 2008 MDVE.NET. Some rights reserved.
This design is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License.

This is the final design.  There were a few mistakes in the prototype, the 1st and 3rd LED digit strobes were switched, a few capacitors that didn’t have the right lead spacing, and the logo was bad… ProtoExpress didn’t silk screen over the hole in the solder mask.  The parts are renumbered and mistakes have been corrected.

Files:frequency_counter.zip (1 MB)

The POV-Ray image has some inaccuracies and missing parts, I’m just learning how to use Eagle 3D.

This circuit takes a square wave signal (0 – Supply Volts) and displays the frequency on six 7-segment LEDs, for a range of 0 – 1MHz.  It counts the pulses occurring in one second, and displays the count over the next second.  It needs a 5-15 volt regulated power source, which will be supplied by the Oscillation Overthruster.  The zip file includes the Eagle PCB CAD files, Gerber files, the bill of materials, a few PNGs of the board and schematic, and where to get the full Creative Commons license.

I tested the circuit using a 5 volt supply, but all the ICs are rated for over 15V, so a regulated 12V supply and square wave input should work just fine.

This was designed using Eagle Light version v5.20 for Linux.

Partlist exported from frequency_counter.sch

I recommend http://www.protoexpress.com for manufacturing.

Accuracy: assuming the crystal passes it’s +/-30ppm rating, that would be:
100% * ( 2^22Hz +/- (2^22Hz * (30 parts/1 MHz)) ) / 2^22Hz , right?
100 * ( 2^22 + ((2^22) * (30/1e6)) ) / 2^22 = 100.003
So that’s +/- 0.003 % maximum error.  Sounds pretty good to me.  I’m not really sure that the ( +/- 30 parts / 1MHz ) constant is the correct way to represent the error factor.  The counter chip also latches on falling edge, but the master reset on level (from what I can tell on the data sheet) so there may be some loss due to the length of the master reset pulse formed by R7 and C5.

Copyright (c) 2008 MDVE.NET. Some rights reserved.
This design is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License.