A new DIY electronic coil tester.

[Myxr6turbo]

Yes, seems as though the new version of the PCB and BOM is different to the original and they had all the parts. The total was $81 then $40 shipping for 5 units fully loaded.

[Matt in California]

Great to hear things are moving along with purchasing the boards! It is good to hear all parts are available. In the earlier version of the board parts that were selected were discontinued. Now out of the tens of thousands of choices the best ones are hopefully chosen.

Regarding the software the following is correct for: the newest version of the board, with the SH1106 OLED and the Nano Every. (I use the Nano Every because it has a higher performance and is cheaper, but a standard Nano should be just fine. I have used generic Nanos with success, but some issues, so I don't recommend it.) This is the hardware that I recommend using. But I hope that you could see where to make the changes near the top of the code after the comments:

//****ATTENTION!!!**** CHANGE HARDWARE VERSION TO YOUR CONFIGURATION!!! ****ATTENTION!!!****
#define Hardware_Version 4 // 1= v1.0x; 2=v2.0x; 3=v3.0x; 4=v4.0x *This changes pin out
#define Display 2 // 1 = SSD1306; 2 = SH1106; 3 = ST7789
#define Bar_Graph_Adjustment 2 // 1 = RPM adjustable; 2 = Center of bars adjustable with potentiometer
#define Micro_Processor 2 // 1 = ATmega328; 2 = ATMega4809(Nano Every); 3 = SAMD21 (Nano 33) //#3 untested

Please let me know if you have any questions.

Matt

[Myxr6turbo]

Thanks Matt, I tried looking for the nano every but couldn't find anything here in Australia.

Hopefully i have the rest of the parts this/next week (just got the V4 boards today). Can't wait to start testing them out. I might 3D print a box to hold everything. Once I'm happy with it I'll post the .STL here.

Thanks again!!!!

[Matt in California]

As long as you use the original nano (not 3.3 version) you should be fine. Just be sure to select to define the microprocessor as 1.
Have fun!

[Myxr6turbo]

Would it be ok to use a model T spark plug instead of 2 pins to create the spark?

[Been Here Before]

Would it be ok to use a model T spark plug instead of 2 pins to create the spark?

Of course.

[speedytinc]

Wouldnt it be better?
Isnt the spark supposed to jump 1/4"?

[Been Here Before]

Wouldnt it be better?
Isnt the spark supposed to jump 1/4"?

1/4 inch is a gap of .25 inch.

Suggested plug gap for a Model T Motor (stock) .031" - .035"

[speedytinc]

Wouldnt it be better?
Isnt the spark supposed to jump 1/4"?

1/4 inch is a gap of .25 inch.

Suggested plug gap for a Model T Motor (stock) .031" - .035"

I understand that, but we're talking about the gap for the tester.
A .035 gap doesnt necessarily mean you have a strong enough coil spark.
I believe the gap for a HCCT is 1/4"

[Been Here Before]

Wouldnt it be better?
Isnt the spark supposed to jump 1/4"?

1/4 inch is a gap of .25 inch.

Suggested plug gap for a Model T Motor (stock) .031" - .035"

I understand that, but we're talking about the gap for the tester.
A .035 gap doesnt necessarily mean you have a strong enough coil spark.
I believe the gap for a HCCT is 1/4"

Depending on your source, it requires about 15,000 volts to jump an air gap of 1/2 to 1 inch. So the 1/4 inch may be the threshold that the coil is delivering sufficient voltage to cause combustion.

[Matt in California]

1/4 inch is a gap of .25 inch.

Suggested plug gap for a Model T Motor (stock) .031" - .035"

I understand that, but we're talking about the gap for the tester.
A .035 gap doesnt necessarily mean you have a strong enough coil spark.
I believe the gap for a HCCT is 1/4"

Depending on your source, it requires about 15,000 volts to jump an air gap of 1/2 to 1 inch. So the 1/4 inch may be the threshold that the coil is delivering sufficient voltage to cause combustion.

The reason that you want a 1/4 inch gap is because that is about equivalent to the gap when the spark plug is under pressure in the engine. Your alternative is to create a pressure chamber for the sparkplug as you will find on some spark plug testers.

Happy Fording!

Matt

[Been Here Before]

I understand that, but we're talking about the gap for the tester.
A .035 gap doesnt necessarily mean you have a strong enough coil spark.
I believe the gap for a HCCT is 1/4"

Depending on your source, it requires about 15,000 volts to jump an air gap of 1/2 to 1 inch. So the 1/4 inch may be the threshold that the coil is delivering sufficient voltage to cause combustion.

The reason that you want a 1/4 inch gap is because that is about equivalent to the gap when the spark plug is under pressure in the engine. Your alternative is to create a pressure chamber for the sparkplug as you will find on some spark plug testers.

Happy Fording!

Matt

"In a 4:1 compression engine, the voltage required to make a spark (at TDC) goes up 4 times. For example, a perfect 4:1 engine (neglecting advance) with a plug gap of 0.025 of an inch requires 1,900 volts (1.9kV) at no pressure and 7,600 volts (7.6kV) at pressure to create a spark."
https://www.gasenginemagazine.com/equip ... z22djzols/

For discussion: "The T engine displaces 2.9 liters (177 cubic inches), and has a bore of 3.75 inches and a 4-inch stroke. The compression ratio was a mere 4.5:1, creating normal cylinder pressures of only 50-75 psi."
https://www.enginelabs.com/engine-tech/ ... d-model-t/

[Myxr6turbo]

Not sure what's going on. The coils were buzzing in the car but I can't get them to buzz with the tester. I replaced the capacitor in one of them so it's showing a pass but I cannot get any of them to spark. Any ideas? What tests can I do? I have tried with a 12v battery and the 6v (9.4v) T battery.

[Luke]

Shawn,

I've never used one of Matt's boards so hopefully he'll turn up shortly to give some advice, however if you're able to read capacitance it suggests that at least part of the coil-tester connection is likely to be ok.

While we're waiting for Matt; it may be that the tester isn't switching properly (for example the FET could be faulty), if it were me I'd check that out with a 'scope, or possibly a suitable multimeter. Do you have either of these things?

Luke.

[Matt in California]

Not sure what's going on. The coils were buzzing in the car but I can't get them to buzz with the tester. I replaced the capacitor in one of them so it's showing a pass but I cannot get any of them to spark. Any ideas? What tests can I do? I have tried with a 12v battery and the 6v (9.4v) T battery.

Shawn,
The joy of troubleshooting at a distance! From what you say, I assume the controller powers up and you see info on the display.

Spark LED
There are three LEDs: +5VDC, +12VDC and a Spark LED. The Spark LED should blink every time the Arduino sends out the pulse, so you should see it blink every time spark should be released. If there is a blinking spark LED then the issue is in the power circuit. Perhaps the Power is too low.

Power Supply
You should have 12 VDC input. If it is under 8 VDC the MOSFET will not work. What type of battery are you using?

After I hear, back on this we can discuss what next.

Matt

[Myxr6turbo]

Thanks Matt for the reply. No the spark LED isn't flashing. The other 2 LED's are on. I used my 12v 700A car battery. I have 5 FACT boards so I ordered a new Arduino and screen but same problem. I'm looking to buy a 240VAC to 12VDC power supply. How many Amps does the FACT need to operate?

Thanks again,

[Matt in California]

Thanks Matt for the reply. No the spark LED isn't flashing. The other 2 LED's are on. I used my 12v 700A car battery. I have 5 FACT boards so I ordered a new Arduino and screen but same problem. I'm looking to buy a 240VAC to 12VDC power supply. How many Amps does the FACT need to operate?

Thanks again,

Shawn,
The unit doesn't use much current. I have found small 12 VDC 7Ah lead cell batteries are more than enough. I have also used various AC/DC adapters...

The two power LED show that you have voltage at 12/5VDC. You should see the third LED partially hidden under the display when you press SW2.

It looks like we have need to focus on the digital side of the circuit. From your earlier post I assumed that you were able to test capacitance, thus the Arduino/display are working, but your last post states otherwise.

When you load the program on the Arduino, do you get error messages?

What version of Nano are you using?

When I hear more, I hope to narrow down the issue.

Matt

[Myxr6turbo]

Great news. I have it working. If I plug in a 5v charger to the arduino and 12v into the FACT it works! Thanks again Matt for helping out. Now I need to read up how to fix the double spark and I'll have 1 good coil

[Matt in California]

Lovely! I have a video of the tool I use to adjust the coils:
https://youtu.be/CWCJyA8rR4A?feature=shared
This tool works for both the upper and lower bridge.

I have made some updates since the video on this tool for the upper bridge adjustment, but it seems to be MIA at the moment.

Here is a quick video showing how to use the tester:
https://youtu.be/LrtVXkEvRSw?feature=shared

I have used the unit that way, but it is designed to work off 12VDC. The good thing is that you can test coils with 6VDC in this configuration to get them ready for hand crank starts.

That said, I believe you may have an issue with either the +5VDC going into your Arduino or a issue with the regulator on the Arduino.

What version of Arduino did you have? I have used a generic units the original Arduino Nano and the Arduino Every. What recently I have focused on the Arduino Every.

Matt

[Jswope]

I finally got my tester up and running. Thanks to Matt I would have never gotten it done without his help. It works fantastic. I replaced all the capacitors and points on all my coils and 3 out of the 4 worked perfectly. The froth on has a bad primary. Thanks again to everyone involved in this project and a special thanks to Matt for all of his patience and helping me get to this point!

[BuzzBumbleBee]

Currently starting a build of a FACT coil tester, my aim is to make 2 new versions.

1) generic board with the MOSFET driver + current sensing + capacitor sensing components that can be connected to any 5V MCU (picture below)

2) an "all in one" PCB that has an STM32 TSOP20 package that should have an even lower BOM total than using a nano (and should have better ADC resolution and sampling rates)

Very early draft of the generic board

[Luke]

Shane,

Good to see some life still in this project, it will be interesting to follow your build.

Are you planning on making the boards and/or design available to all?

Luke.

[BuzzBumbleBee]

Shane,

Good to see some life still in this project, it will be interesting to follow your build.

Are you planning on making the boards and/or design available to all?

Luke.

Absolutely, once it's tested and confirmed working, all schematics and firmware code will be available here / GitHub.

[Luke]

Excellent

[BuzzBumbleBee]

PCB all made up, all waiting to assemble it inside the casing

[BuzzBumbleBee]

Ignore the low current reading (I have the 30amp sensor and forgot to update the configuration) it actually is around 5.5amp peak.


But is this an example of a double spark situation?

[Luke]

Seems no-one else is going to respond ... if you have a look at page 2 of this thread you'll see various images showing different coil responses as they appear on a 'scope and/or OLED display. Hopefully you'll find something useful there wrt to what you're seeing?

Will be interested to see your integrated design. Some time ago I briefly stuck something together using a Nextion screen which allowed interaction and control of operating parameters, along with a much better voltage/current sensor, so there are certainly further improvements that can be made.

Luke.

[BuzzBumbleBee]

Luke,

Thanks for the reply, my plan is to get a full set of coils tuned in at 2ms TTF on the known working device.

Then I'm going to move to a STM32 MCU (48Mhz) with USB support. It should give the following advantage.

- Faster ADC reads at around 10-15 micro seconds (currently the audrino implementations takes around 25 microseconds) should give better sampling resolution.
- Better time tracking +/- 1 microsecond using a 1Mhz internal timer.

==============

However I think the current Arduino implementation has already shown why the car I have access to ran poorly.

All 4 coils had a double spark and non where the same TTF, fastest was 1.8ms, slowest was 2.4ms

[Matt in California]

But is this an example of a double spark situation?
PXL_20240515_153502047.jpg

I didn't double check, but I thought I had posted the manual here. Here is the FACT Cheat Sheet:

For myself I really find the Oscilloscope function, either single fire or continuous, to be the most useful. I hope this is helpful.

Here is a pdf of the manual.

Model T FACT Adjustment.pdf

(3.45 MiB) Downloaded 113 times

I have not invested much time on the forum lately. If you have questions feel free to send me a message using email. Perhaps we could have a phone conversation, just include your phone number and when to call/time zone.

Enjoy,
Matt

[BuzzBumbleBee]

Thanks for the info Matt, I was able to get all 4 coils running very well - all 2.0ms (+/- 0.05ms)

I have ordered a 2nd revision of my coil driver board and a small STM32 based breakout board (see below)

These should be with me in the next few days, hopefully I'll have a working STM32 version of the firmware in a week or two.

[BuzzBumbleBee]

Here is the schematic for my PCB, it should be compatible with both 3.3 and 5v logic (with the placement of 1 extra LED)

I have also included the kicad files if anyone wants to make there own or modify what i have made

[Lawt]

This Topic has blown me away with it’s enthusiasm and pooling of talent for the betterment of life with a Model T. Every contribution (nearly) has been positive and supportive but worthy of a special mention is Luke for the initial concept and code development and obviously Matt for turning it into a beautifully packaged PCB with a graphic display. Awesome guys. As a “T” newbie (4 weeks) this has also been incredibly informative to help understand how the coils actually work and how they should be set up for optimal performance. I’m no stranger to Arduino’s but I still learnt some new techniques around fast scanning - as they say every day is a school day!

Matt - I received my V4 PCBs from JLCPCB yesterday and had it fully built up and running perfectly within a couple of hours. My coils are now all tuned up to perfection having found one double sparking, one firing rather late and another firing very early, today I had my longest adventure to date (only 25 miles) but she ran a dream.

I bought the minimum order of 5 PCBs and had all 5 fully populated. I want to keep one as a spare but if anyone in the UK wants to benefit from these guys talent then please let me know.

[Luke]

Tim,

Appreciate the comments, thank you, and good to see you've got your unit operational so quickly, I expect Matt would be pleased to read that.

Jolly decent of you to make the offer of boards to others in your country too.

Luke.

[Matt in California]

Matt - I received my V4 PCBs from JLCPCB yesterday and had it fully built up and running perfectly within a couple of hours. My coils are now all tuned up to perfection having found one double sparking, one firing rather late and another firing very early, today I had my longest adventure to date (only 25 miles) but she ran a dream.

Tim,
Thanks for your post! It has been a while since I took the time to read much on the forum let alone build some boards. I always appreciate to hear from others. Great to hear that the process was smooth for you. It has been a great international collaboration! I appreciated Luke's original ideas and everyone else up to you posting your experience. I has always been my hope that someone like you would take the project on and build a handful and share with others. It takes a bit of skill to soldier and program, so it is not for everyone.

Feel free to personal message me with any thoughts or questions.

Matthew

[Alaric]

I'm trying to order some PCBs and have run into an unavailable component:

Straight-through L2,L4,L3,L1 BEAD-TH_BD3.5-L6.0-P10.00-D0.9 4 ML3.5*6.0*0.8 Meled(??) LCSC C46030 0.008

Can anyone suggest a suitable substitute that they would have in stock?

Thanks,

Richard

[Matt in California]

Richard,
Yes, that is an issue. Interesting that the part I selected they seem to removed all information!

Try replacing with part# C5339819
https://www.lcsc.com/products/Ferrite-Beads_527.html

Let me know if you have any other questions.
Matt

[Alaric]

Matt,

Thanks for the new part number.
I have 10 boards ordered and was able to order the Ferrite Beads from the link you provided.

Richard

[kernelpanic]

Hi,

I just wanted to mention that the tester works great! I put it together as a small winter project and used it to adjust my coils. Big thanks to all of you, it's a fantastic project!

[Matt in California]

Moritz,
Thanks for your feedback! Feel free to share about your process. Did you send the boards out to be made? Did you have any trouble programming?

Matt

[kernelpanic]

Matt,

I ordered the PCBs from pcbway and the components from LCSC. Since my job has to do with electronics, I soldered the PCB myself. I highly recommend ordering the PCBs pre-assembled for anyone else. Without soldering experience and a hot air soldering station, it’s quite difficult.

As for the programming, Arduino is not new to me, so it wasn’t a problem.
I just unfortunately didn’t notice that many displays have VCC and GND swapped; one display had to suffer because of that

I replaced the capacitors and the points of my coils and then set them to 2ms. Only one coil had double sparking; the others were very easy to adjust, just time-consuming.

I also designed and 3D printed an enclosure for the tester:
https://www.printables.com/model/123849 ... oil-tester

Moritz