Model Engine Ignitions

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Well I'm slowly finding time to make some model engine ignitions and for my race car. I am developing several different types of ignitions I am hoping to sell here, my web page, and on other sites. Once I have the designs ironed out I'll get a vendor membership for here and sell them. But in the mean time I would like some feedback on the designs. Can't promise to incorporate them but I'll try. So follow along and see what happens, R&D research and destruction.

I'll be offering the ignitions in 3 versions, just the board, full kit, and preassembled versions. No I don't have a final price yet but kits should range from $30 to less than $100 dollars depending on features and model. The versions range from a simple buzz coil to fully programmable sequential ignition timing. I have learned from manufacturing other equipment, mostly industrial stuff like hardware and software, that it is hard to please everyone. But that does not mean that I can't offer some choices. I spent most of the winter designing these boards to allow one to choose to use certain parts by leaving them out and just bypassing the part, I'll explain more later. I would like to say that I may appear to be new here but I'm not. I'm also not new to the internal combustion process (deflagration). Yes I'm also fully aware that there are others offering ignitions and there are lots of DIY'ers out there but this is just my offering. I am also a very firm believer in the "K.I.S.S. System". I tried to make the boards user friendly for soldering. The way I see it, if I find it easy to solder them then others should be ok, if not then I'll redesign the boards. The boards can either use my custom designed coils or you can use your own, just run wires to your coil.

The boards, a work in progress:
I used new software to design these boards "Dip Trace" it is the best I have used so far for the price and I have used software from Win-Draft to OrCAD, so $50 to $5,000 or more.
The first one below is my Buzz coil "IDI-ADJ-BZ-HV" or Inductive Discharge Ignition, Adjustable, Buzz, High Voltage". I did research about Buzz Box ignitions and found that there was timed and untimed boxes. I can't see why anyone would want to use untimed today. The "IDI-CD4047B" or Inductive Discharge Ignition using a CD4047B which is a one-shot multi-vibrator that operates from 3.5v to 18v. Both the IDI-ADJ-BZ-HV and the IDI-CD4047B use the CD4047B set for non retriggerable so that it can't refire until the current trigger pulse is done. This also means that if the engine is stalled at a trigger point the coil current shuts off after the pulse. The CD4047B needs to see a rising or falling pulse in for a trigger. The ignition coils I had made were based on an inductive coil that I had them modify so it can work in IDI or CDI. I don't like the term TCI because maybe I won't use a transistor.

Features for Buzz:
-Operates from 4.7v to 18v, higher the voltage the stronger the spark. But normally set for 5v using a reg for above 6v in.
-Auto Advance with RPM, if you want. Selectable (fixed or auto) or leave out the switch and hard wire a jumper.
-Use either points, inductive, or Hall-Effect trigger (min 3.5v trigger source).
-Max auto advance is controlled by the distance between the Hall-Effect to magnet.
-Use trailing edge with points, inductive, or Hall-Effect to get auto advance.
-Use my coil or yours.
-Adjustable number of sparks and at what frequency, this is so you can use your own coil, each coil is different. The dwell time R3 is the window for the sparks, frequency R7 is the actual number of sparks.
-Self resetting fuses.
-3" (76.26mm) long x 2" (50.49mm) wide.
Buzz
Buzz Board Web.jpg


Features for IDI CD4047B:
-Same as the buzz coil board but puts out just one spark. Less parts also.
-Adjustable dwell time R3.
-2.75" (69.79mm) long x 1.75" (44.38mm) wide
IDI CD4047B
IDI CD4047B Board WEB.jpg


Features for CDI Sparky-1:
I originally made this board to work with R/C plane engines and it works very well. With 11.7 volts in it will make sparks up to 37,200 RPM on a single cylinder operation, 21,000 RPM at 5v operation. Average HV is 27,000 or 18,000 at 5v and 32,000 at 11.7v.
-Capacitor Discharge Ignition.
-As low as 400ma current consumption, can go lower but the spark is weak.
-Uses an external trigger and power source. I'll probably change the external regulated power source to internal.
-Needs a 5v external trigger in. Doesn't matter what type but must be at least 5 volts.
-Voltage in is 5-12 volts. I have ran it with 4.5 volts.
-Uses a hic-up circuit so that once the capacitor is charged it turns the charging circuit off and on as needed to keep the capacitor charged and reduce power consumption.
-Charing voltage is 1,000v at 1,000 RPM (single) and 100v at max RPM, max RPM depends on voltage supply.
-Has a separate Kill Switch in line.
-Sparky-2 and 3 in development. Less parts and better performance I hope.
-3" (76.43mm) long x 2" (50.73mm) wide.
Vid Prototype development 1
Vid Prototype development 2
Sparky-1
Sparky-1 Board Web.jpg


Features for Universal Trigger Board CD4047B:
I designed the Universal Trigger Board to put out a signal from just about anything. I also designed it work with small engines to automotive use.
-It is a signal conditioner. Signal in square wave signal out of 5-15 volts.
-Will work with points, inductive, or Hall-Effect trigger, has built in weak single boost op-amp.
-Use the Hall setting for points and Hall-Effect.
-Use HEI setting for inductive coil weak signal setups.
-Use with Sparky-1 board.
-Fixed or Auto-Advance
-2.75" (69.85mm) long x 1.75"(44.45mm) wide.
Universal Trigger Board CD4047B
Universal Trigger Board Web.jpg


Well these are my simplest boards but, I could have made something as simple as the TIM-6 board but that to me is too simple. But wait, you can use either the Buzz board or the IDI-CD4047B board as simple as the TIM-6 board it's just a matter of choosing what you want and bypassing everything else. For example: if your using a 6v battery you can jumper F1, jumper the input pin of the voltage regulator to the output pin, if you want just fixed timing then leave out the switch and jumper pin 3 or 4 of the switch to pin 2 of the switch, for auto-advance same thing but jumper to pin 5 instead of 2, and so on. These board designs are not locked and I only ordered 5 of each as prototypes for R&D purposes. If someone want's to try one contact me, just pay me for the board and shipping, I'll probably sell 3 of each and any extra parts I have like coils and such. Oh the boards already have a spot for the pull-up resistor for the Hall-Effect so it should be just a matter of putting a 1.5k ohm resistor in and connecting a Hall. Oh the regulator is mostly for the Hall-Effect depending on which one you choose.

How the auto-advance works:
The auto-advance works by using the trailing edge (negative going) of the input signal. It doesn't matter if you are using either points, inductive, or Hall-Effect trigger, as RPM goes up the trailing edge will move closer to the leading edge and so timing will advance from it's initial position. For points you put a 10k ohm resistor in for R1, put a 10k ohm resistor between SIG-IN and S-, then put the points across S+ and SIG-IN. This will give a high when the points close and a negative going low for the auto-advance. You can't control the max advance with points (yet) but you can with a Hall or inductive pickup. When you move the Hall-Effect or inductive pickup away from the magnet the whole pulse width will shrink. So you just reset the initial timing and the auto-advance will now advance less.

Current research:
Currently I'm working on a programmable ignition timing interface. Using an Arduino Nano the interface will allow me to use either points, inductive (HEI), or Hall-Effect trigger working off the crank or distributor signal to program 200 points at 50 RPM increments. The PITI is to replace the mechanical advance on my race car, I don't use a vacuum advance, to much power for that. The interface goes between let's say the dizzy and the ignition box. In my case a Crane High 7 ignition box. The interface is similar to the UT board but, programmable with all the bells and whistles I want and need for drag racing. It's also scalable up or down. Eventually it will do sequential ignition firing from 2 to 20 cylinders or more if I want but, that's getting ahead of myself and I have my racecar to get ready.

Anyway let me know what you people think or if you have any questions just ask, I'll get back ASAIC.

Ray
 

petertha

Well-Known Member
HMEM Supporting Member
Joined
Jun 24, 2010
Messages
1,906
Reaction score
601
Interesting. Can you show some examples of the RC installations you did? Were they single cylinder or multi? I'm slowly finishing a methanol glow radial. Gasoline/spark would have been the preference but at least at my level it was opening up yet another can of worms. I've seen smaller more scale like plugs and I've seen some conversion ignition kits but I haven't delved too deep. Plus I'm electrically 'challenged' Lol.
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Interesting. Can you show some examples of the RC installations you did? Were they single cylinder or multi? I'm slowly finishing a methanol glow radial. Gasoline/spark would have been the preference but at least at my level it was opening up yet another can of worms. I've seen smaller more scale like plugs and I've seen some conversion ignition kits but I haven't delved too deep. Plus I'm electrically 'challenged' Lol.
Unfortunately because of health reasons I had to give up working on the ignitions 3 years ago and I'm just picking it back up. This all started years ago on RC Groups and I was part of the group working on a project called "Programmable-Open-Source-CD-Ignition-PIC1840". We did have running engines and I had purchased some engines for testing myself but I didn't make it there. This stuff here in this thread is pretty simple and straight forward.

One day I hope to make a radial. If your going to use a distributor then an ignition would be fairly easy but, if you want to go distributor-less then that would require 1 ignition coil for each jug. Or if you want to use a dizzy just for spark distribution that would be ok also. For my Olds 455 I'm using a 5.0 litre (302) cam sensor assembly (replaces the dizzy), crank sensors (4 in a aluminum pulley), and 8 near-plug coils. I haven't built that ignition yet, I wanted to do these other ones first. Track is closed right now because of covid-19. For your engine I would wait until I get my programmable ignition for small engines done, I'm working on that right now. This way you can tell the micro-controller where your cam and crank sensors are and it takes care of the rest, you just need to set your ignition curve after that. By where I mean how many degrees BTDC they are. Sync pulse would most likely be for # one. I have seen where people have tried to use just 1 coil and electronically distribute the spark but, that didn't work out very good.

As for examples; I will be doing a Tiger 26cc 2 stroke, a 2 stroke OS60, and once it is rebuilt an ASP 65 Four Stroke, all singles. I'll post videos to my You Tube channel when they are done. My question is 'where would you get the cam timing from on a scale radial'?

Ray
 

petertha

Well-Known Member
HMEM Supporting Member
Joined
Jun 24, 2010
Messages
1,906
Reaction score
601
Re cam timing you would either have to measure it directly from the engine with the usual protractor wheel observing TDC & rocker/valve lift. Or get the IO,IC,EO,EC info from manufacturer specs (assuming they reference it, which often they don't). Or in the case of engines from plans, reverse engineer it as I had to do. But from ignition perspective, I've seen links where people have converted methanol glow plug engines to gasoline spark ignition & basically referenced spark X degrees off TDC. The cam timing just is what it is. Like most things, I suspect there there is likely variation & tuning relative to the engine itself. I'm also not clear if/how advance/retard is somehow integrated into these solid state systems, haven't got that far myself.

Popular conversion reference

Example gasoline multi-cylinder engines with corresponding ignition modules. Some work better than others form what I've gathered. Another issue specific to RC is EM noise.
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Yah I guess I should have explained better. I could do the same setup as CH Ignitions which is a wasted spark setup for radials, that way there is no need for a cam sensor or dizzy. The sync pulse for #1 cylinder comes off the same rotating hub as the crank position signals. One has to remember that micro controllers (MCU) and micro-processors (MPU) can not predict the future and they need time to do the calculations and look-ups. So the magnets need to be placed say 45 - 90 degrees before the firing time it all depends on how much timing advance you want and MCU calculation time. To put more advance into timing you put less delay in firing. I'm sure you are aware that some engine builders want true scaled down parts and are willing to spend the time and money to get that look. If true looks is not a problem then the below setup is the way to go. In the picture below from CH Ignitions I drew an arrow pointing out the sync magnet for cylinder #1. This setup below is the easiest way to do it. The black magnet is #1 for lining up the firing order and the magnet before it is there to tell the MCU that the next pulse is going to be #1 cylinder. This can be done either by timing the pulses or by using 2 Hall-Effects, say N-S for cylinders and one S-N for syncing #1 cylinder. In this case 8 magnets are used but, one could also just use one magnet and sense notches or teeth to do this as in automotive. For my Olds 455 I'm going with 8 coils for the cool look LOL. Converting a glow engine to spark is fairly straight forward for fixed timing or simple auto-advance but, a programmable timing one is more difficult. As for cost between a single output one and the one shown below is 7 coils instead of 1 and about 16 more micro chips, still need those plug wires. As for EMI/EMF, if weight is not a problem then that is easy to take of.

CHI.png


Ray
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Well I would probably purchase one to try. I have a 9 cylinder radial about half done and will need an ignition system. I also have a v8 85 percent done so 37k spm sounds good to me.
For a really good spark I need 0.0025 seconds (2.5 milli-sec) between sparks for a good spark which means your radial could do 24,000 RPM. For the V8 which requires 4 sparks per rev would top out between 10,000 - 40,000 RPM using 1 coil, with 4 coils it would top out at 40,000 RPM using 12 volt supply. The coils I use require 1.5 milli-sec to charge and 1 milli-sec to dis-charge which limits the RPM. Because with the CDI I charge up to 1,000 volts it has a really good spark at idle, good for anti-fouling. The charge voltage goes down as RPM goes up and this happens with all CDI's no matter who makes them. Most CDI's will work down to @100-150 volts on the primary and the coils are usually 100:1 winding ratio so output on the secondary drops down to 10-15,000 volts. My coils are 200:1 ratio but that does not mean that my secondary output is 200,000 volts because CDI ignitions don't work that way. The impedance of the capacitor and coil really knocks things down.
I usually try to get someone I don't know personally to test a system so long as they agree to give me honest detailed feedback and understand that they would be working with a beta-prototype (prerelease system), I'll keep your hat in the ring if you want. I should have testing samples by the end of August.

Ray
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
So why am I making these boards/systems? Well it's because I need them for myself and I figure others could use them also. Now I'm a bit of a perfectionist which means I'm usually never satisfied with something until I can abuse the sh_t out of it and it survives R&D. I've been fascinated by and studying IC engines for 46 years now. From 020 Peewee engines to blown alcohol engines to 20 cylinder EMD 645 diesel engines, I still want to make a scale SD40-2. So far most of the electronics I make is for stuff I use/need or stuff I invent and license to companies to make and sell. Well that's enough about me.

I've been checking the boards I have shown above, usual QC stuff because the boards came on a slow boat from China and I'm using new software (Dip Trace) for designing boards. I already don't like the clearances around solder points and I have to increase the clearance, otherwise one would need a fine tip soldering iron and steady hand. Other than that they seem ok for now.

None of the boards use surface mount parts which adds to the cost but, not everyone can solder surface mount parts. The Sparky-1 board uses a SMD transformer to up the voltage for the CD but, the tabs are big enough and spaced enough for easy soldering. You will need fine resin/rosin core solder to solder the boards.

I'll be providing YouTube videos and resource material on my website as work progresses.

Ray
 

bluejets

Well-Known Member
Joined
Apr 22, 2014
Messages
422
Reaction score
147
Hi Ray,
One of the biggest problems with the 1840 and earlier was it didn't seem to matter what effort most put in, there was always a request for something different. Seemed no one was satisfied no matter how it was presented.
That combined with, as I remember it, the hurdle of getting enough energy for high speed ignition into the cap in the short period of time.
Poor old Gompy near pulled his hair out on many occasions.
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Hi Ray,
One of the biggest problems with the 1840 and earlier was it didn't seem to matter what effort most put in, there was always a request for something different. Seemed no one was satisfied no matter how it was presented.
That combined with, as I remember it, the hurdle of getting enough energy for high speed ignition into the cap in the short period of time.
Poor old Gompy near pulled his hair out on many occasions.
I remember all that. I got ahold of Microchip for help with the PIC1619 but, they didn't seam to care about that chip. They also didn't seam to know much about the chip other than the basic resources so I started looking into other options. I really didn't want to take a lead on the 1840 because I was busy with my computer business. Now that I'm semi retired I have more time to play around I guess one could say. I started looking around for a processor and IDE that was more or less for beginners and chose the Arduino Nano. Early testing with the ATmega328P shows it only needs 5 - 7 clocks per instruction which is nice. I haven't tested the FPU on it yet but, I feel confident in it. I'm just going to mount the Nano boards straight onto a mother/daughter setup. I have learned C# and have been making GUIs that work with the Nano USB for programming variables and the EEPROM lookup tables on a Nano.

I'm not going to try and satisfy everyone's wishes. Small changes, yes, big or custom no. Yah it's to bad Gompy couldn't make ago of things. I know how hard it can be running a small business. I had to give up my computer support business, the stress gave me permeant heart problems, so I go slower now and enjoy my grandkids.

Take care
Ray
Some of my test boards
Board Samples.jpg
 

Willyb

Well-Known Member
Joined
Jan 21, 2016
Messages
57
Reaction score
9
Hi Ray
I would certainly be interested in helping you out any way I can. Using the Arduino Nano is a good way to go. It will be nice to have another ignition system available for our model engines.
Cheers
Willy
 

bluejets

Well-Known Member
Joined
Apr 22, 2014
Messages
422
Reaction score
147
Hi Ray,
You might want to give some thought also to using the Arduino pro mini and programming it through a usbasp programmer.
I have found it to be more positive programming result than normal especially if using some clone boards.
Easy to do and with the pro mini a lot samller installation.
Video below is about custom bootloader but process for standard code is the same.

 

mccreamark

Member
Joined
Jun 13, 2013
Messages
6
Reaction score
2
++ for Arduino here and the pro mini especially for battery powered projects, I use them regularly at home and at work (R&D engineer). If you need more speed or sophistication you can go for the Mega or Due same IDE. I also use the Texas Instruments Launchpad boards on the Energia IDE which is a clone of the Arduino IDE. They have an array of processors and Fram for storing data.
 

bluejets

Well-Known Member
Joined
Apr 22, 2014
Messages
422
Reaction score
147
Speed for size etc. etc. etc............ is as Ray already showed in the ESP32.
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Thanks bluejets & mccreamark about the Pro Mini. I also looked at the Pro Micro which uses the ATmega32U4 which is nice and it should work. It has the 0 &1 ex-int which I need and of course the PCINTs. I to have been looking at the TI line and I'm committed to using the STM32 line with ARM M4 and the M0 cores for a major project but those would be overkill for these little projects. STM is the only manufacturer that I could find that had an RTOS that is UL certified and that took me 9 months to get it from them. My only concern with the Pro Mini is the lack of a native USB port for uploading the parameters. Mind you at this point just about everything can be changed. The PIC12F1840 was an experience in being conservative and maximizing resources. My ignition controller interface uses the following:
• 3 adjustable rev limiters in 50 rpm increments
o 1 for burn out use a switch
o 1 for staging – foot brake (switch) or trans brake
o 1 for Max rpm
• Adjustable timing in 50 rpm increments to 10,000
• Adjustable timing by the tranny gear- use switches (3)
• Trick shift timing retard by switch or rpm in 0.001 sec. increments (1)
• Start up retard ignition timing by 1 degree increments – use starter solenoid signal (1)
• Hall effect 5v – 20v, 12 volt signal out. Universal trigger board (RPMin pin 2)
• GM style HEI signal compatible op-amp (RPMin pin 2)
• Water/Engine temp (pin 10)
• Uses 5 Inputs ?

I'm not sure if a 16 MHz 8-bit CPU can do what I need but I have to weigh flexibility to cost. I'll go up in CPU power and features as required. Eventually I will combine everything into a stand-a-lone sequential distributor-less ignition. We have the sequential distributor-less ignition but it is not programmable. It can however hook up with an MSD programmable ignition controller but now the cost jumps up to $1,100 USD, ouch.

Anyway by using the Arduino IDE and either the Nano, Mini, or Micro the cost stays down and is easy to learn. Hopefully I can have something running this weekend. Oh I either have to fix my timing light or make one (Gompy style) :)

Ray
 

Vietti

Well-Known Member
HMEM Supporting Member
Joined
Dec 8, 2020
Messages
215
Reaction score
99
Location
Thermopolis Wyoming USA
Admit I know next to nothing about building one of these, I'm a magneto guy. I have S&S ignitions that work great. A feature to consider is making the timing light standard, very handy, and I have wanted the cheap plug in tach offered on the RCEXL ignitions, can you do it?

John
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Admit I know next to nothing about building one of these, I'm a magneto guy. I have S&S ignitions that work great. A feature to consider is making the timing light standard, very handy, and I have wanted the cheap plug in tach offered on the RCEXL ignitions, can you do it?

John
I have the timing/tach light on all the ignitions and I could make it so one could use the RCEXL one or add something in. Speaking of RCEXL I have sourced out the same parts that they use such as the Hall-Effect holder and sparkplug caps & wire. I am also working on a programmable Krober Magneto CDI but it has proven to be a challenge, CPU keeps resetting. Firstly I need to get the basic stuff nailed down and get some testing done before moving on.

Ray
 

Willyb

Well-Known Member
Joined
Jan 21, 2016
Messages
57
Reaction score
9
Hi Ray
I would second the addition of a simple Tach like the one used with the Rcexl Ignition systems. To make it less work for you, why not just use the Rcexl one.
Looking forward to this project.
Cheers
willy
 

74Sprint

Well-Known Member
HMEM Supporting Member
Joined
Apr 21, 2013
Messages
251
Reaction score
63
Location
Winnipeg, Canada
Well my saga continues. Normally I would not discuss any project that is in the works but, I feel this time I want people to see the amount of work it takes to make quality parts. Because I'm not designing something custom I have to give the end user some options. For example to replace the simple TIM ignition with something more modern one can use the CD4047B demo circuit below.

Ignition Using CD4047B.pdf

Schematic Ignition Using CD4047B.png

Board Top Layer.pdf
Board Bottom Layer.pdf

If you want anyone can play with the circuit. R1 & C1 control how long the pulse width is which controls the dwell for the best spark. The nice thing about using a 1 shot is it can be set to non-retriggerable which means it can't be re-fired until it is finished firing the present pulse. Also to say if the engine stops turning and the points should be closed or the Hall-Effect stops on the magnet the power to the coil shuts off after the pulse is done and won't turn back on until the next rev.

Anyway, because I'm using new software (Dip Trace) and boards from a new supplier you never know what your going to get until you get the boards. So as I mentioned I don't like the clearances around the solder pads and now I'm finding some of the holes are too small, more changes. The holes are plated through holes and I find the holes in Dip Trace are not allowing for the plating so they are about 0.001"-0.002" too small. Also my pattern for my spark coils are off a bit, I need to redesign the pattern. As you can see the pins are at an angle, also the holes are a bit tight.

Tight Coil.jpg

I also have to order some parts from Digi Key. I thought I had them but nope. Anyway, I'll get this Buzz Board working soon.

Ray
 

Attachments

  • Ignition Using CD4047B.pdf
    97.1 KB · Views: 301
  • Board Bottom Layer.pdf
    77.2 KB · Views: 217
  • Board Top Layer.pdf
    77.4 KB · Views: 233
Top