Ignition system for Farm Boy

[bluejets]

I have purchased a 4 pin dc scooter CDI system and a scooter coil. I anticipated the same issues of using a standard hall sensor and not being able to trigger the CDI unit.

I did a write up on this same approach......
Model engine CDI easy and cheap

 

[Bernett1]

I saw your other post but the hall sensor I purchased is different than the one you used. I’m hoping that because it’s an inverted sensor that I can eliminate the transistor circuit you added. I’m just wondering if you think the inverted hall sensor will work? Thanks

 

[Bernett1]

I’m new to this group. I don’t have the expertise or equipment to build engines but I really love the work this group does. I am using converted 2 stroke engines for RC planes and wanted a less expensive CDI system than the rexcl systems. I have purchased a 4 pin dc scooter CDI system and a scooter coil. I anticipated the same issues of using a standard hall sensor and not being able to trigger the CDI unit. I did some searching and located some inverted hall sensors. There aren’t many of them. I was able to get the dc CDI unit to spark using the inverted hall sensor. They are so small that mounting them is a challenge.

 

[simonbirt]

A quick update on the ignition circuit. I have built one on a bread board and it worked fine. A bit temperamental as the connections are not that good. Next step is to make one up on using proper joints, I may even have a go at designing a PCB with one of the free cloud based apps and ordering a couple of boards. I am using a 6V motorcycle coil, could perhaps do with something a bit smaller but it is what I could find.

 

[Tim Wescott]

A couple of suggestions from a professional circuit designer:

First, check to see if you need heat-sinking on that TO-220 case transistor. The best way to do this is probably to run the circuit at full speed or more, under load (i.e., with a spark plug), for a few minutes. Then shut it down and feel the tab on the thing -- carefully. If it's too hot to touch, it needs a heatsink. If you're left with a transistor-shaped blister, you weren't careful enough!

The rule of thumb is that if you can hold your thumb on it, it's less than 50 degrees C. If you're in a typical 25C ambient room, that means 25C of temperature rise, which should make the circuit good for operating in a hot engine bay.

Second, if you're in the US, consider buying your boards from OshPark -- they're a very nice circuit board aggregator. I don't have recommendations if you're outside the US -- I'd do a web search on "circuit board aggregator" and see what popped up. If you're someplace that's in tight with the EU there are some really good ones out of the former Soviet Bloc; if you're somewhere that's in tight with Asia there's good ones in India and China (there's also some appallingly bad Chinese ones -- so ask around if you go there). You'll need to check to make sure that your online tool will generate Gerber files. Or you can just get KiCad or Eagle and learn a whole new set of skills .

Make your tracks wide enough. The default track width for any program is going to be pretty small -- this is a power circuit, so you want to think about how much current the tracks will be carrying and size them accordingly. Probably anything that's not associated with the collector or emitter of that power transistor will be fine at the default trace width, but the current through that transistor is going to be significant. There's circuit board track size calculators out there that you can use -- or just make sure that anything that's in the path from the final transistor's collector to ground is 1/8" wide and call it good. If it burns up, make the traces wider on the next board!

 

[bluejets]

A quick update on the ignition circuit. I have built one on a bread board and it worked fine. A bit temperamental as the connections are not that good. Next step is to make one up on using proper joints, I may even have a go at designing a PCB with one of the free cloud based apps and ordering a couple of boards. I am using a 6V motorcycle coil, could perhaps do with something a bit smaller but it is what I could find.

Good for you....and as you've already found, breadboards are not the best especially for any device pulling some decent current.
As for the board design, take a peek at Doc's submissions for some idea as his works fine.

Youtube video here.......

 

[simonbirt]

A couple of suggestions from a professional circuit designer:

First, check to see if you need heat-sinking on that TO-220 case transistor. The best way to do this is probably to run the circuit at full speed or more, under load (i.e., with a spark plug), for a few minutes. Then shut it down and feel the tab on the thing -- carefully. If it's too hot to touch, it needs a heatsink. If you're left with a transistor-shaped blister, you weren't careful enough!

The rule of thumb is that if you can hold your thumb on it, it's less than 50 degrees C. If you're in a typical 25C ambient room, that means 25C of temperature rise, which should make the circuit good for operating in a hot engine bay.

Second, if you're in the US, consider buying your boards from OshPark -- they're a very nice circuit board aggregator. I don't have recommendations if you're outside the US -- I'd do a web search on "circuit board aggregator" and see what popped up. If you're someplace that's in tight with the EU there are some really good ones out of the former Soviet Bloc; if you're somewhere that's in tight with Asia there's good ones in India and China (there's also some appallingly bad Chinese ones -- so ask around if you go there). You'll need to check to make sure that your online tool will generate Gerber files. Or you can just get KiCad or Eagle and learn a whole new set of skills .

Make your tracks wide enough. The default track width for any program is going to be pretty small -- this is a power circuit, so you want to think about how much current the tracks will be carrying and size them accordingly. Probably anything that's not associated with the collector or emitter of that power transistor will be fine at the default trace width, but the current through that transistor is going to be significant. There's circuit board track size calculators out there that you can use -- or just make sure that anything that's in the path from the final transistor's collector to ground is 1/8" wide and call it good. If it burns up, make the traces wider on the next board!

Thanks all useful stuff.

Simon

 

[AndrewW]

Looking really good Simon.
Cheers
Andrew

 

[Mago]

Attached is a photo of a circuit board I milled on my cnc machine It is 70mm by 50mm so large enough for fat fingers are able to do the wiring with ease.
It takes "normal" electronics and soldering up is no problem.
The other attachment is the circuit. I did this some 10 years ago but I don't expect much has changed since.
I have used it on a number of models.
When using this type of ignition using a magnet always be aware that it won't work if the incorrect pole is presented to the pickup.
I always mark the correct face of the magnet so it shows after fixing. 3mm magnets shave a habit of reversing pole without warning..
The circuit drawing as below.

Mago


file:///C:/Users/user/Documents/6v%20Transistor%20ignition%20circuit..docx

 

[L98fiero]

file:///C:/Users/user/Documents/6v%20Transistor%20ignition%20circuit..docx

The link didn't work.

 

[Tim Wescott]

The link didn't work.

That ended up being a link to the file on Mago's computer -- which will work fine for Mago. It just won't work for the rest of us.

file:///C:/Users/user/Documents/6v%20Transistor%20ignition%20circuit..docx

You need to use the "Attach files" button, below and at the left side of the screen where you enter text.

 

[simonbirt]

As promised here is an update on the ignition module. I have made a short video as it shows the beast in action.

 

[AndrewW]

Hi Simon
You've made a very professional job of that. Very impressive!
I'm looking forward to trying it out on the Jerry Howell Hit n Miss engine I'm currently building.
Cheers
Andrew

 

[Tim Wescott]

Excellent job for a first-time PCB layout. In fact, it's pretty good leaving out the fact it's your first. I have two nits for you, if you ever do a board turn.

First:

While you did good with nice generous trace widths on the power path, it wouldn't hurt to bump up the default tracewidth for everything else. It looks like you're using about .01"; you could go up to 0.02" or possibly even 0.05" -- basically, just make them as wide as you can without losing convenience.

Second:

With all that room on the board you could put a big copper ground pour on the top side to the left of the power transistor, bend the legs, and screw it to the board for a heat sink. Solder mask is epoxy, so you could probably leave off any insulating pad and just use thermal paste, or you could use a pad for convenience. Then if someone did need a heat sink, it'd be built in. You'd crowd the other components a bit, but you've got acres of extra space on that board.

 

[simonbirt]

Excellent job for a first-time PCB layout. In fact, it's pretty good leaving out the fact it's your first. I have two nits for you, if you ever do a board turn.

First:

While you did good with nice generous trace widths on the power path, it wouldn't hurt to bump up the default tracewidth for everything else. It looks like you're using about .01"; you could go up to 0.02" or possibly even 0.05" -- basically, just make them as wide as you can without losing convenience.

Second:

With all that room on the board you could put a big copper ground pour on the top side to the left of the power transistor, bend the legs, and screw it to the board for a heat sink. Solder mask is epoxy, so you could probably leave off any insulating pad and just use thermal paste, or you could use a pad for convenience. Then if someone did need a heat sink, it'd be built in. You'd crowd the other components a bit, but you've got acres of extra space on that board.

Thanks Tim for the advice, not really my forte but I thought I could do with learning a new skill. For some reason I've had plenty of time at home of late. If there is a V2 then I will incorporate the ideas, I will also put the diode to protect the LED on the board, I only discovered I needed it after I had ordered them.

 

[bluejets]

Well done Simon.
The old original drawing probably caused a bit of confusion so I drew up what I think should be a clearer drawing.
I only redrew the section in question as the rest remains as is, except for your added zener diode.
What value did you use here...? I'm guessing almost any would do around 6v2..??

I tend to use the small pre-made boards these days off Ebay (KY-003).
Anyhow, drawing below shows the connections with the 2 hall effect in series, Q3 being the normal ignition timing and Q4 (if so desired) on the closed exhaust for a hit and miss engine.
Points are self explanitary so no need to redo that.
Operation is fairly simple in that both are required to be closed for the unit to fire.
Remembering here that closed points are for firing. (opposite to kettering system)

Cheers Jorgo

 

[Rudy from Norway]

Thought I could throw in my own designed Farm Boy ignition circuit. It is a "Buzz Box" type. It makes several sparks for a given time. So if used on a higher reving engine the number of sparks will decrease as the revs increase. Haven't tested it on higer reving engines though.
I'm a pro hardware development engineer, but I'm not all that familiar with the quarks of the 555 (I use a dual 555, the NE556). I would like a second opinion on the design. The case is that even if it works grate at 6V, it shuts down on 6,5V or more. Something is not optimal.
Maybe you Tim Wesckott has a tip?
Here you can see it running:
Rudy

 

[simonbirt]

Just goes to show how many ways there are to skin a cat.
You have also given me an idea for the base, I like the fact that it is all self contained with everything hidden.

 

[simonbirt]

Well done Simon.
The old original drawing probably caused a bit of confusion so I drew up what I think should be a clearer drawing.
I only redrew the section in question as the rest remains as is, except for your added zener diode.
What value did you use here...? I'm guessing almost any would do around 6v2..??

I tend to use the small pre-made boards these days off Ebay (KY-003).
Anyhow, drawing below shows the connections with the 2 hall effect in series, Q3 being the normal ignition timing and Q4 (if so desired) on the closed exhaust for a hit and miss engine.
Points are self explanitary so no need to redo that.
Operation is fairly simple in that both are required to be closed for the unit to fire.
Remembering here that closed points are for firing. (opposite to kettering system)

Cheers Jorgo

View attachment 121984

Well done Simon.
The old original drawing probably caused a bit of confusion so I drew up what I think should be a clearer drawing.
I only redrew the section in question as the rest remains as is, except for your added zener diode.
What value did you use here...? I'm guessing almost any would do around 6v2..??

I tend to use the small pre-made boards these days off Ebay (KY-003).
Anyhow, drawing below shows the connections with the 2 hall effect in series, Q3 being the normal ignition timing and Q4 (if so desired) on the closed exhaust for a hit and miss engine.
Points are self explanitary so no need to redo that.
Operation is fairly simple in that both are required to be closed for the unit to fire.
Remembering here that closed points are for firing. (opposite to kettering system)

Cheers Jorgo

View attachment 121984

The diode is just a 1N4003 (what I had) seems to do the trick. I only have a limited understanding of electronics so rely on people who know better. The diode inclusion was worked out from first principles and some help from the web.

Regards,

Simon

 

[bluejets]

Rudy....the ignition Simon is using is a buzz unit also with cut out if no pulse detected.

Simon...you show the diode in series with your LED as a zener diode but quote as a 1N4003 which is a standard diode.(D1)
Connection you show as a zener i.e. reverse which is correct but a standard diode will not conduct in that configuration in normal operation.
Do you have a clear photo of your completed circuit and of the finished pc board, just to clarify what you have done.

Cheers Jorgo