Ignition circuit help

[canadianhorsepower]

This is the latest.

I admire your effort and courage. if people would only stick there finger up ANY EMPTY HOLE they wouldn't side track someone asking for info:wall:

 

[dsage]

Looks good Steve:

Give it a try. Even with your oldschool car coil first just in case the coil you want to use is bad.

(I try to do my best despite the distractions )

Sage

 

[stevehuckss396]

Looks good Steve:

Give it a try. Even with your old coil first just in case the coil you want to use is bad.

Sage

Okie Dokie!

 

[canadianhorsepower]

simply want o apologize for some bad expression I have use. I guess some expression can be use in a shop but not on a public site
Steve: If I can make any sugestion, simply purchasse a small dc motor (12volts) at radio shack and mount a small plastic plate
plate at the end then put a screw and a smal magnet 180 degree apart. You can control the speed with a small potentiometer
to simulate a distributor. I use this all the time to check circuit

 

[dsage]

Canadianhorsepower:

Good idea with the motor and magnet thing. This way you can use the actual magnet that will be used on the engine. (see below)
I've always triggered such circuits with a 555 timer oscillator but with that method you don't get the variablel dwell (circuit activation time) that you would get with the spinning magnet idea. I'll have to try that. Thanks.

On that note:
It was pointed out earlier (by someone else) that the TIM circuits with a hall sensor do not operate exactly like a conventional points type ignition. With the TIM circuit, the coil is normally de-energized, then energized for a short period of time while the magnet is in front of the sensor and then the circuit turns off again.
The important point is that the spark occurs when the sensor is cleared not when it is activated. Therefore there will be a built in error in your spark timing by the width of your magnet. As long as you have adjustment in your timing (usually by turning the distributor) and you use a timing light this poses no problem.

The other thing is the magnet size becomes somewhat important in that at the highest RPM you wish to operate, the sensor (and the coil) must be activated for a reasonable amount of time to "charge up" properly. A couple of milliseconds should be enough and you can calculate that by observing how many degrees of rotation the magnet activates the hall sensor and figuring the activation time from deg per second at the maximum RPM. If you're a bit short just mount a slightly bigger magnet.

In my V8 the magnets are 3mm in the distributor (equiv of 6mm on the crank) so they don't need to be large. But worth some rough calculation to be sure you don't run out of spark at high RPM's.

Sage

 

[canadianhorsepower]

[/With the TIM circuit, the coil is normally de-energized, then energized for a short period of time while the magnet is in front of the sensor and then the circuit turns off againQUOTE]

That's wright,but quick fix to this.:idea: I'm using a op amp (741) and use the input from the hall sensor and connect it to the inverting
input pin (2) then my output (6) match the circuit. I also use a 4mm rare earth magnet . These are strong enought that i dont have to mount
the sensor real close to the magnet.
As for the timing marks I was turning my motor by hand and with a led where the coil goes I could see that the spark happen wright after
the magnet pass the sensor and that work 100%

 

[dsage]

How did you arrange for the split +/- power supply for the 741 opamp?
Actually, I suppose it doesn't matter in this application that the output of the 741 won't swing all the way to zero volts when you operate it from a single supply. I guess it would work ok.

The big issue with inverting the logic so the coil is ON most of the time (instead of off) is that when it's ON it's drawing maybe 15amps. Your battery won't last long. And a lot of things get hot. It's more power efficient to have the circuit off most of the time especially on a two cylinder engine where it spends most of the time waiting for the next spark anyway. On an eight cylinder there becomes an issue at high RPM's not having enough time for the coil to "charge" up between sparks so it's just as well to leave it on.


Sage

 

[canadianhorsepower]

How did you arrange for the split +/- power supply for the 741 opamp?

I don't understand this one???

The big issue with inverting the logic so the coil is ON most of the time (instead of off) is that when it's ON it's drawing maybe 15amps.

your right about this one but your only going to generate a spark when you cut the magnetic field.
If we do the math for an 8 cyl at 4000 rpm = 16,000 pulse per minutes time 60 for seconds and we geta frequency of96000 hz
I'd say that at that speed is has to be on some time to recharge

 

[dsage]

The 741 opamp requires a + supply (above ground) and a minus supply (below groound). So you'd need a minus supply for it to work properly. If you run it on a single supply (12v and ground) the opamp output will not go to zero volts. Instead it will only swing to (I've forgotten) but some low value above zero. As mentioned it probably makes little difference in this circuit. Although there are single supply opamps available and you could use one of those instead (forgot the part number).

But, a much simpler approach would be to add, yet another transistor and resistor to make an inverter on the front end. Much easier to lay out on a circuit board as well instead of that 8 pin DIP device.

Check your math. You have the 16,000 sparks per minute correct but that's only 16000/60 = 267 sparks per second, or one spark every 4ms.

As I mentioned it does become an issue for an 8 cylinder engine but for a 2 cylinder it's only 67 sparks per second or one every 15ms.

A two cylinder would not be in the same predicament until it reaches 16k RPM.

Sage

 

[dsage]

Steve:

Looking back a bit I see I missed your one liner asking about the IGBT transistor.
I goofed up my drawing part number (sort of)
The correct part number is an IRGB14C40LPBF not IRGS....
Same part but the IRGB is a TO220 tabbed variant capable of being screwed down.
I guess too many numbers and letters to get straight in my head.

Having said that for a low power application the tab(less) device would probably run cool enough to survive right on the board.

I updated the drawing in the previous post.

Thanks

Sage

 

[canadianhorsepower]

Come on [QUOTE]The 741 opamp requires a + supply (above ground) and a minus supply (below groound). So you'd need a minus supply for it to work properly. If you run it on a single supply (12v and ground) the opamp output will not go to zero volts[/QUOTE]
If you dont have the below ground it will go to zero but not below ground. I can show you a picture from the scope to show it.

[QUOTE]But, a much simpler approach would be to add, yet another transistor and resistor to make an inverter on the front end. Much easier to lay out on a circuit board as well instead of that 8 pin DIP device[/QUOTE]
it's your choice, but single transistor are a bit old school the trend now is op amp. I'm not using 741 cause the opamp I'm using is a 4 channel for my injection system and I'm also using it in a comparator mode

Check your math. You have the 16,000 sparks per minute correct but that's only 16000/60 = 267 sparks per second, or one spark every 4ms.

no no no my math are fine Revolution Per Minute will always be 60 time less then revolution per seconds
their is 3600 seconds in one hour so take your RPM multiply it by 60 to get Hz so 16k Multiply by 60 will be 96khz

 

[stevehuckss396]

Steve:

Looking back a bit I see I missed your one liner asking about the IGBT transistor.
I goofed up my drawing part number (sort of)
The correct part number is an IRGB14C40LPBF not IRGS....
Same part but the IRGB is a TO220 tabbed variant capable of being screwed down.
I guess too many numbers and letters to get straight in my head.

Having said that for a low power application the tab(less) device would probably run cool enough to survive right on the board.

I updated the drawing in the previous post.

Thanks

Sage

Roger That!

I am going to build the version 2.1 circuit with the device I have on hand. I am hoping to get the board done Sunday and do some testing. Should have some news Sunday night if I dont have to work.

 

[dsage]

>>> their is 3600 seconds in one hour so take your RPM multiply it by 60 to get Hz so 16k Multiply by 60 will be 96khz

Not sure how HOURS started figuring into it. We were talking about revolutions PER MINUTE and sparks per MINUTE and Hz - sparks per second.

If you have 16,000 sparks in a minute you have 1/60th that many in a second (there are 60 seconds in a minute)

Do you actually think the engine puts out 96,000 sparks per second ?? Think about it. (the definition of Hz is per second)

16,000 sparks per minute / 60 seconds in one mnute = 16,000 / 60 = 267 sparks per second.

Figure it backwards.

267 sparks per second x 60 seconds per minute = 267 x 60 = 16,000 sparks in one minute.


Let's look at it yet another way
The engine is rotating at 4000 RPM that's 67 revolutions per second
By your calculation - where's the engine going to get rid of 96000 sparks in 67 revolutions.

Sage

 

[dsage]

Ok Steve:

Let me know if something doesn't look right. I'll be glad to help you work through it.

Sage

 

[Mosey]

Roger That!

I am going to build the version 2.1 circuit with the device I have on hand. I am hoping to get the board done Sunday and do some testing. Should have some news Sunday night if I dont have to work.

Of course you are building an extra for me aren't you?

Hang int here, Buddy, you're doing great!

Mosey
(Got my TIM-6 today)

 

[cfellows]

Really interesting thread. Steve, have you calculated how much the parts, excluding circuit board, will cost for this?

Also, would be interested in knowing the maximum current this circuit can supply to the coil.

Finally, would this circuit work with points or a micro switch instead of a hall effect device?

Thx...
Chuck

 

[stevehuckss396]

Hello Chuck!

The components cost me $11.81 excluding the board. $14.31 including the board. Because the resistors come in a 5 pack, if you make 5 of them it would be $7.04 each or $8.54 with the board. The board is a blank 4 X 6 inch that I had to cut on the mill with an engraving bit.

The output transistor is rated 15 amps with the proper heat sink.

I believe it would work with a switch from the output to the negitive. If we can get Dsage or CHP to verify that I will redraw the circuit with the switch in it for future reference.

 

[canadianhorsepower]

I believe it would work with a switch from the output to the negitive. If we can get Dsage or CHP to verify that I will redraw the circuit with the switch in it for future reference

Hi Steve,
sure it will work but I'm a little bit confuse why add points condensor and wires, or switch and wires when all you need is a magnet
now one question about your project why are your magnet spread 135 apart???????????

 

[canadianhorsepower]

can someon tell me how to remove to yellow question marks that keep showing on

 

[stevehuckss396]

Hi Steve,
sure it will work but I'm a little bit confuse why add points condensor and wires, or switch and wires when all you need is a magnet
now one question about your project why are your magnet spread 135 apart???????????

Some folks might already have a model with points in it and want to go to an electronic ignition. It would be an option for repairing or replacing an existing ignition.

The magnets are 135 degrees apart because the combustion cycle of the engine is 270 degrees apart.

Pop - 270 - pop - 450 - pop - 270 - pop