Van's Air Force

The definitive Van's Aircraft support community! Buying, building or flying an RV? Join our exclusive family of mentors and enthusiasts!

PMagger 2.0 - Ignition Control for the price of a burrito........

petehowell

Well Known Member
Patron
As of late I have been experimenting with ignition monitors. I finally have one that checks all my boxes. I have a hybrid ignition system on my plane. The right side has a Megajolt/EDiS system that was fun to make and test with the help of all the smart people here - I have several hundred trouble free hours on it now. The left side is a vintage 20+ yr old P-Mag (PM)that has been updated along the way and has given great service for almost 3300 hrs now. The MegaJolt (MJ) gave me one thing the P-Mag never really could - Control. I like to run LOP and my carbed O-320 needs a fair bit of advance to do that smoothly.

Big note of caution here - your engine is probably different and doing advance wrong can ruin your expensive engine

The MJ has a switch that allows moving b/t 2 independent timing curves - this is great for 2 situations 1) ROP climbs - lower advance (20- 24 deg) really helps keeps temps under control and 2) LOP cruise where more advance (30-40 deg) helps light off a slower burning lean mixture. The P-Mag has ground adjustable timing that in my experience is good for one or the other(for my engine).

Now for about the price of lunch at the burrito store - you can control the P-Mag and get the best of both worlds. More caveats - this is very experimental - there are commercial products that do the same - if you want a real product and support call them. if you like to build cool stuff - read on.... I have bench and flight tested this, but you will want to do the same. This is not a product - it is an idea, and a starting point. Be very careful here. All that said, it works pretty darn well.

What is it - It's an ignition monitor that also has a P-Mag control element. It's based on an ESP32 microcontroller, a small TFT screen, and an RS232 to TTL converter. You can buy that hardware for about $15. Code was written for ESP-IDF 5.4. It displays timing advance, timing correction(shift), ignition state (takeoff or cruise) and RPM for both the MJ and the PM. It has a switch input that lets you flip between 2 max advance limits on the PM. It wirelessly streams advance data for logging (I stream it to my GFT EFIS via another ESP32 to join engine and flight data). It also sets up a wireless hotspot you can log into and change P-Mag parameters on the ground eliminating the need for the sometimes futzy EiCad.

Less interesting to most people here is that the ESP32 acts as a USB host to ingest data from the MJ - that was the most challenging bit of code to get working. It still suffers data timeouts, but the recovery is pretty graceful. My coding skills are 40 years old and I'm not that smart - so Claude helped me(as did a few great guys here on the forums). I do finally have (nerdalert!) a Github if anyone is interested. Modifying this for a dual PMag setup would be pretty easy. I have no idea if this is of interest to anyone, but I enjoyed the process and the results.

My current setup for my carbed O-320 running mogas is: Curve shift = 5.6 deg, Takeoff mode = 21 deg advance limit, Cruise mode = 40 deg advance limit(but I only see about 35).

I take off ROP and climb to altitude - level off and start leaning the engine - I know at WOT and ~2250 rpm I am LOP at about 6GPH - when I get there, the engine will be a bit rough - flip the switches and the engine gets smooth. I have found that the advance gives me smooth ops and about 3kts TAS running LOP over the reduced timing. 3kts is great, but the big plus is the cooler climbs.

Big thanks to all the smart guys here that have done so much to educate us on ignition dynamics - Dan, Larry, Toolbuilder, Nigel and many others. My plane gained some utility here. If you are curious - try it, if you think i am crazy, my wife agrees with you :)

Here is a video of it in action.


Here is the web interface

AP1GczNs_UUMmkbXAdxFYE87-w4dR1v11rteXLpzYHIH2KWNw6qzh5jhVx4-6bdDGifs7dAWNNDBJFk7Nh_0k9Iqacs5uEIXkl8Wxr1H8y0sGPYarYXl0prt-fHWnmyPLSG4MNrSouSM7YWss2sxp7k0RpLPGw=w438


Here is how put it in the panel for now

AP1GczP19KtF2NZnDc6kdl7VZvPhOUPfVCHSMEB77ctADDOxapIb4UUcAEkgPbKYpSWHoyEmkTiZEijcOt1VW-edl1oAozVJNENTmNbQ7E6bL2dwW5WHn0g7rQoCzYXKEnpTYCZ6h2joWKWTAwmsdFIKiXkrVw=w1000


Flight test engineer with the post flight celly!
AP1GczNehqKaDW78IW7gQqUkdFJA2nL_Fly343XqhJNEbaoh941SFDdsmbSWcRz8QB606EGVvvqW4FRTf0jCOIP2DYC5_YSIh0mUm0PkX8F0E6IYEQsMfnJmYtl8z-fLI90zE45_V6DuzHgChFDR70Yf52bBmA=w1000
 
As of late I have been experimenting with ignition monitors. I finally have one that checks all my boxes. I have a hybrid ignition system on my plane. The right side has a Megajolt/EDiS system that was fun to make and test with the help of all the smart people here - I have several hundred trouble free hours on it now. The left side is a vintage 20+ yr old P-Mag (PM)that has been updated along the way and has given great service for almost 3300 hrs now. The MegaJolt (MJ) gave me one thing the P-Mag never really could - Control. I like to run LOP and my carbed O-320 needs a fair bit of advance to do that smoothly.

Big note of caution here - your engine is probably different and doing advance wrong can ruin your expensive engine

The MJ has a switch that allows moving b/t 2 independent timing curves - this is great for 2 situations 1) ROP climbs - lower advance (20- 24 deg) really helps keeps temps under control and 2) LOP cruise where more advance (30-40 deg) helps light off a slower burning lean mixture. The P-Mag has ground adjustable timing that in my experience is good for one or the other(for my engine).

Now for about the price of lunch at the burrito store - you can control the P-Mag and get the best of both worlds. More caveats - this is very experimental - there are commercial products that do the same - if you want a real product and support call them. if you like to build cool stuff - read on.... I have bench and flight tested this, but you will want to do the same. This is not a product - it is an idea, and a starting point. Be very careful here. All that said, it works pretty darn well.

What is it - It's an ignition monitor that also has a P-Mag control element. It's based on an ESP32 microcontroller, a small TFT screen, and an RS232 to TTL converter. You can buy that hardware for about $15. Code was written for ESP-IDF 5.4. It displays timing advance, timing correction(shift), ignition state (takeoff or cruise) and RPM for both the MJ and the PM. It has a switch input that lets you flip between 2 max advance limits on the PM. It wirelessly streams advance data for logging (I stream it to my GFT EFIS via another ESP32 to join engine and flight data). It also sets up a wireless hotspot you can log into and change P-Mag parameters on the ground eliminating the need for the sometimes futzy EiCad.

Less interesting to most people here is that the ESP32 acts as a USB host to ingest data from the MJ - that was the most challenging bit of code to get working. It still suffers data timeouts, but the recovery is pretty graceful. My coding skills are 40 years old and I'm not that smart - so Claude helped me(as did a few great guys here on the forums). I do finally have (nerdalert!) a Github if anyone is interested. Modifying this for a dual PMag setup would be pretty easy. I have no idea if this is of interest to anyone, but I enjoyed the process and the results.

My current setup for my carbed O-320 running mogas is: Curve shift = 5.6 deg, Takeoff mode = 21 deg advance limit, Cruise mode = 40 deg advance limit(but I only see about 35).

I take off ROP and climb to altitude - level off and start leaning the engine - I know at WOT and ~2250 rpm I am LOP at about 6GPH - when I get there, the engine will be a bit rough - flip the switches and the engine gets smooth. I have found that the advance gives me smooth ops and about 3kts TAS running LOP over the reduced timing. 3kts is great, but the big plus is the cooler climbs.

Big thanks to all the smart guys here that have done so much to educate us on ignition dynamics - Dan, Larry, Toolbuilder, Nigel and many others. My plane gained some utility here. If you are curious - try it, if you think i am crazy, my wife agrees with you :)

Here is a video of it in action.


Here is the web interface

AP1GczNs_UUMmkbXAdxFYE87-w4dR1v11rteXLpzYHIH2KWNw6qzh5jhVx4-6bdDGifs7dAWNNDBJFk7Nh_0k9Iqacs5uEIXkl8Wxr1H8y0sGPYarYXl0prt-fHWnmyPLSG4MNrSouSM7YWss2sxp7k0RpLPGw=w438


Here is how put it in the panel for now

AP1GczP19KtF2NZnDc6kdl7VZvPhOUPfVCHSMEB77ctADDOxapIb4UUcAEkgPbKYpSWHoyEmkTiZEijcOt1VW-edl1oAozVJNENTmNbQ7E6bL2dwW5WHn0g7rQoCzYXKEnpTYCZ6h2joWKWTAwmsdFIKiXkrVw=w1000


Flight test engineer with the post flight celly!
AP1GczNehqKaDW78IW7gQqUkdFJA2nL_Fly343XqhJNEbaoh941SFDdsmbSWcRz8QB606EGVvvqW4FRTf0jCOIP2DYC5_YSIh0mUm0PkX8F0E6IYEQsMfnJmYtl8z-fLI90zE45_V6DuzHgChFDR70Yf52bBmA=w1000
A true experimenter! Great job putting all this together. Maybe some day you can teach me modern coding. Last I touched it, C++ was in vogue;)

BTW, where did you find a pink D10?
 
A true experimenter! Great job putting all this together. Maybe some day you can teach me modern coding. Last I touched it, C++ was in vogue;)

BTW, where did you find a pink D10?
It is an early D10 that was anodized vs painted and has faded I think - it's a novelty at this point, and I'm keeping it!

This is coded in C++! Claude really helped get it working.
 
very cool to see the evolution of this! i've been messing around a bit with the dual pmag set up. I think i have a semi-working prototype on a tiny screen. I would love to see the code changes you've made to get to this stage!
 
very cool to see the evolution of this! i've been messing around a bit with the dual pmag set up. I think i have a semi-working prototype on a tiny screen. I would love to see the code changes you've made to get to this stage!
Here is my github - not sure if i set it up right. Comments appreciated. If anyone builds one and improves it - let me know!

PMagger github
 
Hmmm... Now you have me wondering if I need to hire a similar flight test engineer... ;)

I applaud those who puzzle their way through these seemingly-complex components and just keep going until they have something that works. Well done!
 
Experimental aviation at its best :D
Yes! My RV-9 starboard wing recently lost a battle with my hangarmate’s Ford Ranger but my homebuilt “Pete Howell Special” LED nav lights I built up in 2006 or so are still kicking and definitely getting transferred to the new wingtip! There are not many true homebuilders like Pete around anymore!
 
Yes! My RV-9 starboard wing recently lost a battle with my hangarmate’s Ford Ranger but my homebuilt “Pete Howell Special” LED nav lights I built up in 2006 or so are still kicking and definitely getting transferred to the new wingtip! There are not many true homebuilders like Pete around anymore!
Steve - This makes me so happy! Mine are still chugging along too!

This is Baby Kate as LED light spokesmodel way back when - Kate is now married and almost 30.........

AP1GczNmXVKn94ZAq4aGHM6sg6hGFxQDPk8ayqNsY-JbpVIw4-ynfWw91i1TveOOg6LLpFaQcliVq_4fQ3iVwVNvFzjPBKIhU_ZV6tdy3K_bYocqMxZkX6ybqXv2sf1zr_zs_c4UmVV1ZLajpK5dH-XNIfa1LA=w1024
 
This is an incredible project! I think I'm going to have to order some parts. Any advice on using it with a single Pmag? My personal preference is to have one slick as a backup.
 
This is an incredible project! I think I'm going to have to order some parts. Any advice on using it with a single Pmag? My personal preference is to have one slick as a backup.
Nitro - Ready to go for one PMAG - just strip out or ignore the MegaJolt stuff.
 
Pete, is there a way to tell the Pmagger it is running on a clocked P-mag? Or is it going to display a timing value 5 (or 7) higher than the actual timing?
 
Pete, is there a way to tell the Pmagger it is running on a clocked P-mag? Or is it going to display a timing value 5 (or 7) higher than the actual timing?
Hi Dan - quick and dirty update - something like this? - offset(-2) is in the yellow "ADVANCE" line on the PM side. It is configurable in the UI(purple button). It is saved in non-volatile memory so it survives reboots. The overall advance(big white number) is the sum of the PMag value and the offset value. if clocking is a better term, that is an easy change.

AP1GczM1Ss-Gt-35IHVGi4ZN53kWKUVhTlDHw3nMW_ycPWxC4fF0Gtq-xa6ddUgq4tNCoYTXYz6nIj-5du2-twG_VpJherNB0BwSgOCz4_eNnIGyJtk_oFfme9CL8yIND_atfmhiUcddNGKp2-xViMCl0Rtz7Q=w800

user configurable in the web UI
AP1GczN1SKECvXHKIQTPMkS9-oBv7I69YaLWBk5t498WE4q_f6sTPGTwiCgyoGjeWHugRfGxobWTHImKsCH_gvZE-MKczuU390z3JBP1m2U6CbetZIoiriudXDqkk5XIDvs6fv8QmMSOSizjXQ9nN7-b4xKRyQ=w438-

AP1GczPwCorly7ATPRXg5Cpl8-_DxOKKDX7vZZjp31UgfGeIgw5lUJhRzVTvYXTi0MYW6IlY_eoEI9qR1hb-XO9iMqKnSRwvRGau_6f-oWEHNM4g7oZjnQgROxKplv2BtJgqBFW2N2zfo7HRCcbn6YFYE-Qq5Q=w438
 
I added a feature tonight - it allows you to capture actual timing values for both curves as you fly(depending on what curve you have selected). The values are displayed on a second web page and stored in non-volatile memory, so the matrices can be populated over a number of flights. It will tell you for a given RPM/MAP combo - this is the advance that is being used, taking into account any max advance limits, curve shifts, and physical clocking that is being done. It could help diagnose excessive heat or other issues.

I might get out to test tomorrow, but then our WX goes to crap for a while. Let me know if this would be helpful - I know you could capture the data looking at the screen and scratching on paper, but what fun is that?

Here is a screenshot on the old Iphone.
AP1GczOl5_XPkQgibTsD035efWaKALCPqOVmyOkiGQj8barRT0_WCc9MMTN80vy6t2dgOId2yO4XPGvJjKv0Dg5zHWGKsYArnvNpngjyWeRUThsihj2TZcWdGqeR2RbeiYFxsvWWWgL4IloiL7Rh1Iv2SDUpaA=w438
 
Very interesting Pete, and nice work!

I have the engine bridge controller installed, with advance angle displayed on my Dynon Skyview system. Like you, I'm only able to see roughly 35° of max advance, despite allow settings. I wonder if the base advance needs to be set higher to enable higher cruise advance settings?

The other issue I have is the single advance curve. I timed the P-Mags to TDC, then set base advance at -2°, which has greatly reduced CHT's during climb (keeping them below 400 °F). I could see going a bit lower in summer, but we'll see if that is needed. -2° reports ~25.5°, FYI. The feature I like most about your setup is the ability to change between a climb mode and cruise mode. In climb mode, or for short flights under 5k in altitude, I'm likely looking to limit advance angle to 30° to keep CHT's under control. Having to manually change this on each flight is a pain, relative to flipping a switch.
 
Documenting test results here in case anyone is following along.

Got this data on the bench today - Note the advance shift was set to 8.4 - WAY more than you would ever fly with - just to show interesting numbers, and the TO curve was capped at 21 deg(my drill only spins to 2500rpm). Seems to confirm things are working in the PMag brain the way we are telling it to. This is an easy way to build the timing matrix automagically.

I'm looking fwd to flying this config (with less advance) and matching the timing data with other engine and flight data for some analysis.

One point I am looking into is - at the same displayed timing - the engine is noticeably smoother running on just the MJ vs Just the PM. The 'Jolt has fresh plug wires - I will try to make fresh PM plug wires this weekend to see if that is the issue.

Interesting Note: the cruise values below are really close to what Brantel found years ago(when you adjust for the shift I used). Brian did some great work in this area!

AP1GczNXvObzYrb-l0DpgpkuqozCxgjxU32seqzTO6_ZZTcLdQnOxY-RWEkYQFOLu-0F2ykLpH-1ctFVM-TYaQFT-PcVKqA2ouuj3MNaSP9M2KExSFFMY7zY3nttYViCNhivfQz1KIfHTGIEz4pvlTIA7_I6kA=w525
 
Documenting test results here in case anyone is following along.

Got this data on the bench today - Note the advance shift was set to 8.4 - WAY more than you would ever fly with - just to show interesting numbers, and the TO curve was capped at 21 deg(my drill only spins to 2500rpm). Seems to confirm things are working in the PMag brain the way we are telling it to. This is an easy way to build the timing matrix automagically.

I'm looking fwd to flying this config (with less advance) and matching the timing data with other engine and flight data for some analysis.

One point I am looking into is - at the same displayed timing - the engine is noticeably smoother running on just the MJ vs Just the PM. The 'Jolt has fresh plug wires - I will try to make fresh PM plug wires this weekend to see if that is the issue.

Interesting Note: the cruise values below are really close to what Brantel found years ago(when you adjust for the shift I used). Brian did some great work in this area!

AP1GczNXvObzYrb-l0DpgpkuqozCxgjxU32seqzTO6_ZZTcLdQnOxY-RWEkYQFOLu-0F2ykLpH-1ctFVM-TYaQFT-PcVKqA2ouuj3MNaSP9M2KExSFFMY7zY3nttYViCNhivfQz1KIfHTGIEz4pvlTIA7_I6kA=w525
While i have nothing to back it up, i have heard that the PV engines run better on one of the plugs vs the other; likely combustion dynamics, swirl, etc. very possible this is the difference and not the ei. Could also be timing difference. More advance tends to be smoother than less, with in limits of course. So be sure the advances are identical before comparing smoothness. E.g. i run my 320 at 35* at idle rpm. Much smoother than 25. It tapers back to 22, as i go above 1000. You should experiment now that you have this cool tool.
 
While i have nothing to back it up, i have heard that the PV engines run better on one of the plugs vs the other; likely combustion dynamics, swirl, etc. very possible this is the difference and not the ei. Could also be timing difference. More advance tends to be smoother than less, with in limits of course. So be sure the advances are identical before comparing smoothness. E.g. i run my 320 at 35* at idle rpm. Much smoother than 25. It tapers back to 22, as i go above 1000. You should experiment now that you have this cool tool.
Good thoughts - I will play around a bit - I am not sure the numbers are comparable across the ignitions - might have to get a timing light on them when the WX warms up.
 
Do you have the original Brantel numbers?

One issue with the Pmags-I think the timing curve is basically a fixed 7deg, or maybe 8deg advance when you have a controller connected to the mags-it’s effectively in the A curve. In order to achieve greater advance, one needs to increase the advance shift. Is that what you’re seeing as well?
 
Do you have the original Brantel numbers?

One issue with the Pmags-I think the timing curve is basically a fixed 7deg, or maybe 8deg advance when you have a controller connected to the mags-it’s effectively in the A curve. In order to achieve greater advance, one needs to increase the advance shift. Is that what you’re seeing as well?
It is a curve shift -there is the base A curve and the B curve is base + 5.6 I think. I am happy running the B curve, but limiting overall advance on Takeoff/climb to ~22 deg and them when I go LOP run the higher advance of the B curve. Might want to knock that B curve down on a hot summer day or just run the takeoff curve. Mogas runs LOP on the B curve just fine (in my O-320 - your engine may be different!)

Brantel is here - I'll let him post his data - or give me the OK to do it.
 
It is a curve shift -there is the base A curve and the B curve is base + 5.6 I think. I am happy running the B curve, but limiting overall advance on Takeoff/climb to ~22 deg and them when I go LOP run the higher advance of the B curve. Might want to knock that B curve down on a hot summer day or just run the takeoff curve. Mogas runs LOP on the B curve just fine (in my O-320 - your engine may be different!)

Brantel is here - I'll let him post his data - or give me the OK to do it.
That’s basically what I’m planing on. Currently I have my base timing at 23.8 deg. I’d like to limit the advance to maybe 26.6 deg or so in the summer/down low/ROP, then like you, shift to a B-curve for LOP operations. It appears the engine bridge has the ability to save pre-sets that should allow this behavior.
 
It is a curve shift -there is the base A curve and the B curve is base + 5.6 I think. I am happy running the B curve, but limiting overall advance on Takeoff/climb to ~22 deg and them when I go LOP run the higher advance of the B curve. Might want to knock that B curve down on a hot summer day or just run the takeoff curve. Mogas runs LOP on the B curve just fine (in my O-320 - your engine may be different!)

Brantel is here - I'll let him post his data - or give me the OK to do it.
Feel free Pete. You have the torch at this point! 😀
 
Last edited:
Here is what I have - this B curve has a +4.2 shift - if anyone has different - let us know
View attachment 112438
Non pmag user here. I can’t believe they would put a product out into the world that would deliver 32* of advance at 28”. That is nuts! 39* at 23” is not really sane either. Even crazier if you have dual pmags. No wonder so many people here post about excessive cht issues.
 
Non pmag user here. I can’t believe they would put a product out into the world that would deliver 32* of advance at 28”. That is nuts! 39* at 23” is not really sane either. Even crazier if you have dual pmags. No wonder so many people here post about excessive cht issues.
There is a question as to if the reported values match when the spark actually occurs. On my very crude test setup - it looks like it might be a bit less. That said - when I restrict the advance to 22 for climb, the temps really stay cooler.
 
Back
Top