A new ignition circuit

[canadianhorsepower]

I'm in Pueblo Colorado. Mountain time zone I believe you are in Central Time Zone and that would put you 1 hr ahead of me. Your 6 pm would be my 5 pm.
I have two 1N4003 diodes and one NTE 116 diode. Also three Q1s and two Q2s. Also Have spare capacitors on hand. The only thing I don't have I guess is the LEDs and radio shack doesn't have them.

Ron

That's fine by me is tomorrow 7 your time good, if so get that resistor
a few strip wires and it should work tomorrow

cheers
Luc

 

[ozzie46]

UHH Luc if it's 6 pm your time it will be 5 pm my time ok?
Ron

 

[canadianhorsepower]

UHH Luc if it's 6 pm your time it will be 5 pm my time ok?
Ron

yes so 7pm your time will be 6pm my time

 

[jgedde]

Hey guys, I feel like a heel right now... It has come to my attention that there is an error in sheet 2 of the schematics. The ignition module's external wiring is not correct because the E numbers on the circuit block are in error. Looks as if my design tool assigned numbers the way it wanted and not the way I did... To be clear, the error is only on sheet 2. The circuit diagram on sheet 1 is correct.

It was my intent to have a one to one match between the E point location on the sheet 1 schematic and the E location on the block in sheet 2.

Even worse, I haven't been following this thread closely enough to support those who have been having problems.

I've attached revised diagrams. Hopefully an admin can go back and edit the post with the schematics to point to this post.

I am hoping that any miswiring that happened as a result of the problem didn't break any components and cost money for those who decided to give it a try.

Special thanks to Dsage and CanadianHorsepower for their work!

John

View attachment Sheet1.pdf

View attachment Sheet2.pdf

 

[jgedde]

One more thing about the circuit... Remember that it's designed to prevent the coil overheating or the battery going dead if the engine is stopped with the hall sensor positioned in front of the magnet?

I've heard that guys are having trouble testing the circuit and this feature is to blame. While it's on an engine, and the magnet passes by hall sensor quickly, it's not an issue.

OK, so here's what happens. If you put the magnet right in front of the hall sensor during testing, you'll get a weak spark. This is because the circuit times out thinking the engine is stopped. What happens is the IGBT is turned off slowly and a weak spark is produced.

I will say this... My circuit build up fires a full sized lawnmower plug with a 0.045" gap using the J.E. Howell coil. But, the hall needs to be triggered in a similar fashion to what it sees on a real engine.

To test it, I need to pass a magnet rapidly across the front of the sensor. I've been using a magnet on the end of a small, but somewhat long screwdriver. The magnet is waved past the hall sensor quickly.

Doing this, I get a very strong spark. In fact, I've been zapped a bunch of times by this, and it feels stronger than off a lawnmower engine! Ouch! Is it a GM HEI type zap? No, but those of you who've been caught off guard by an HEI will never forget it!

Luc built the circuit also. He says he puts a fixtured magnet in his drill press and the circuit fires just fine.

So, what to do during testing.... You can either speed up the magnet speed past the hall sensor or temporarily put a short across C2. If C2 remains shorted, the circuit will work fine, but the protection against coil overheating and battery drain will be lost.

I've also found that the coil's polarity makes a big difference. If you still get a weak spark, try swapping the + and - leads to your coil.

John

John

 

[fourstroke]

I have been following this thread with interest as I hope to use this system on my hired hand engine, you can see the progress in the "engines from castings" section
However I am having some difficulty obtaining the Q1 and Q2 components here in the UK. I also dont know what kind of hall sensor I need
Can anybody help me with the part number of a hall sensor and perhaps alternative part numbers for Q1&2
Regards
Dougie

 

[canadianhorsepower]

ok here we are even here in Canada some of these parts number are
not "on the shelf " thing
2n4124 can be replace with 2n3904

as for the 2 others this what the specs are "close enough"

if you go to a GOOD reseller finding a match should be a breeze
good luck

http://pdf1.alldatasheet.com/datasheet-pdf/view/183702/IRF/IRGB14C40LPBF.html

as for the hall sensor, pick a number that will suit your project application
and respect polarity and you'll get a spark:fan:

 

[fourstroke]

Thanks Luc
I dont know anything about hall sensors, when I search I get lots of things that are industrial size
Do the come in different voltages, configurations etc
Thanks Again
Dougie

 

[jgedde]

I've tried several different Hall sensors with this circuit and all worked reasonably well. The clear winner for performance and reliability is the OH090U from Optek. Those things are absolutely bullet proof.

Not being one to follow my own advice, I was grounding my engine with just a alligator clip. Of course, the alligator clip popped off and the spark discharged through the hall sensors I had on there (A1302). They didn't make it....

I replaced them with the OH090U's and the engine ran better and the spark was stronger. Then, the clip came off again. I could actually hear and see a small spark jumping between the engine block and the hall sensor ground lead! After I reconnected the ground, both halls still worked fine! Amazingly robust units.

The OH090S is a no-frills hall sensor with a open-collector output. There is no "chopper stabilization" or other stuff we don't really need. Just a hall element, an amplifier and a comparator.

Here's where I got mine: http://www.digikey.com/product-detail/en/OH090U/365-1001-ND/374779 They are slighly larger than from Honeywell or Allegro. I was able to sand the sides of the OH090U to be the same width as the A1302... I don't recommend this - just saying what I did...

Cheers!
John

 

[jgedde]

ok here we are even here in Canada some of these parts number are
not "on the shelf " thing
2n4124 can be replace with 2n3904

as for the 2 others this what the specs are "close enough"

if you go to a GOOD reseller finding a match should be a breeze
good luck

http://pdf1.alldatasheet.com/datasheet-pdf/view/183702/IRF/IRGB14C40LPBF.html

as for the hall sensor, pick a number that will suit your project application
and respect polarity and you'll get a spark:fan:

Be careful specifying the 2N3904 as a replacement for the 2N4124...

It'll likely work, but the circuit wasn't designed for that unit. The 2N4124 has much higher DC current gain (Hfe or Beta) than the 2N3904, especially operating at a switch. When a transistor is operated as a switch (i.e. in either saturation or cutoff), the DC current gain is dramatically reduced. You often have to dig in the datasheet for this information.

In any event, replacing it with the 3904 will cause more base current to be drawn from the input circuit and will change the duration of the "timeout" feature.

In fact when I was designing the circuit, I originally had the 2N3904 or 2N2222A in mind since they are so easy to get. But, to ensure things would work for others the way they did in my prototype, I rejected those. The DC current gain was marginal - right on the edge of being OK.

Just keep in mind that I designed the circuit around the 2N4124's characteristics. This holds true for the other parts as well. Switching to other parts is done at risk of things not working as expected. It's hard for others to support problems if things are changed that have no verification history.

All that said, many of the parts are not Radio Shack items. But, all are available from Digikey or Digikey Canada (digikey.ca). Apart from the transistors, the LED specified is also crucial. They do more than just light up as a timing indicator... They are also part of the timing.

To all, build the circuit using the parts list and schematics from Dave Sage and I and you'll run less risk of things not working.

This isn't meant as a dig at Luc. Sometimes my writing style can be abrasive without intending it... Luc has done excellent work supporting this circuit, especially since I haven't kept up with this thread as well as I should have.

John

 

[jgedde]

UK source for 2N4124

http://www.digikey.co.uk/product-se...ducts/transistors-bjt-single/1376376?k=2n4124

John

 

[fourstroke]

Thank you for the part number of the hall sensor and for the links
Unfortunately digikey and mouser who advertise as UK based ship all products from the US
This means shipping four times the cost of components and VAT to pay when its delivered
I have managed to find all the correct parts in the UK other than the hall sensor but I will keep trying
Thanks for the help
Dougie

 

[canadianhorsepower]

Be careful specifying the 2N3904 as a replacement for the 2N4124...

I've got 5 circuits running with 3904
and no problem yet:hDe:

 

[fourstroke]

Success
I managed to source all the parts I needed here in the UK
I did a bit of soldering today and after one or two rewiring of mistakes I got a big fat spark from a full size plug
I tested the setup with a magnet in a battery drill running at 1700 rpm and no problems
It looks like this setup will be perfect for my model hired hand engine
Thanks to all who took the time to reply to my questions
Regards
Dougie

 

[fourstroke]

One question about the circuit
The plug fires when the led comes on, is this correct?
Thanks
Dougie

 

[canadianhorsepower]

yes , this led serve as a test circuitThm:Thm:

 

[jgedde]

One question about the circuit
The plug fires when the led comes on, is this correct?
Thanks
Dougie

Yes - sort of. When you set your timing, bear in mind that the plug fires when the light just turns off.

John

 

[fourstroke]

Thats what I thought but my circuit fires when the light comes on
If I move the magnet towards the hall sensor the led comes on and the plug fires, the led stays on till I move the magnet away but the plug dosnt fire when it turns off
Dougie

 

[jgedde]

Thats what I thought but my circuit fires when the light comes on
If I move the magnet towards the hall sensor the led comes on and the plug fires, the led stays on till I move the magnet away but the plug dosnt fire when it turns off
Dougie

The circuit has protection against coil overheating or battery drain if the engine is stopped with the hall sensor triggering the coil. After a very short delay - likely imperceptible - the circuit will "lock-out" and fire the plug. It will seem as if the spark occurred when the magnet is applied but it's really happening a short time after. When this happens, the circuit has already discharged the coil and no spark will result when the magnet is removed.

In other words, the circuit fires the plug whenever the first of two conditions occur:
1) Hall sensor is removed from the magnet.
2) A fixed delay time of approximately 20 milliseconds.

When the engine is actually running, the trigger magnet will pass by hall sensor fast enough such that the lock-out circuit will not activate (condition 1 is satisified). In this case the plug will fire just as the light goes off. So, the transition from light-on to light-off is the reference point from which ignition timing should be set.

I ended up changing the 1 uF capacitor in my circuit to 6.8 uF. This increased the lock-out inhibit time and makes the delay much more obvious along with when the plug will fire under normal conditions. 6.8 uF is a standard value but is less common. That said, a 4.7 uF capacitor would be a reasonable choice as well and much easier to get. Most 4.7uF caps you can readily get will be polarized. IN this circuit, the (-) pin goes to the hall sensor input side.

Cheers!
John

 

[ozzie46]

Finally got my engine to run for a little bit. ( N0 fault of the circuit. Had other issues.)


Battery wire came loose as I was trying to start the engine.


Now when I turn on the switch the LED is on constantly but dim when I turn the engine over the LED flashes brighter but plugs don't fire.

What do you think is the problem? Do you think I could have fried the Hall sensor or what?

Help!!


Ron