Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Directly Measuring Ignition Voltage

Status
Not open for further replies.

3v0

Coop Build Coordinator
Forum Supporter
I have been playing with small engines and have been using the 'old school' neon bulb with a resistor to detect the ignition voltage while cranking the engine by hand. It would be nice to have a meter with a digital readout or a graph. Is there any reason not do a voltage divider and sample the voltage with a PIC ?

20K volts max would be enough for these engines. So for a 5V PIC 20000V/5V = 4000 so a 4000:1 divider. The maximum recommended impedance for analog sources is 10 k so that should work.

Suggestions?

3v0
 
Get an inductive pick-up like they used in the old engine analyzers.
 
The duration of the ignition pulse is very short, so unless you can find a way to synchronize the PIC's acquisition period with the trigger event, you're nor likely to catch it. And even that might be rather iffy unless the duration of the pulse is longer than the acquisition window..

I would try using a digital scope with an appropriately sized input voltage divider.
 
These would all be easy answers if one had digital scope and was willing to use it in a not so clean shop.

And tanks Chris you are correct about the speed issue.

Found this which gives me some idea of the main spark duration which is about 100 us but it is very spike-ish.

https://www.globaldenso.com/en/products/aftermarket/plug/basic_knowledge/spark/index.html#

p1-4_1.gif


  1. When the primary current is cut off at the 'a' point, the secondary voltage rises.
  2. At the 'b' point, partway through the rise in voltage, the spark plug reaches the discharge voltage and a spark is generated between the electrodes.
  3. Between 'b' and 'c' is called the capacitance spark. At the start of the discharge, the spark is generated by the electrical energy stored in the secondary circuit. The current is large but the duration is short.
  4. Between 'c' and 'd' is called the inductance spark. The spark is generated by the electromagnetic energy of the coil. The current is small but the duration is long. From the 'c' point, the discharge is continued for about 1 millisecond and at the 'd' point, the discharge ends.
C to D is said to be 1 ms so given that B to C should be 1/10 of 1 ms or 100 us.

My conclusion is that this is doable but impractical unless one were developing a product. Still it could be fun to use a fast ADC and driven by a zylinx programable part fast enough to capture the digital data. For about $20 one can get the MAX1499 which is a 105 MSPs.

It might be an excuse to learn to program PLDs but the high frequencies is what would worry me.
 
Last edited:
Hi 3vo,
as the primary and secondary mirror each other, you could tap the LT side instead.
The connection would be easier to make and would eliminate any potential for misfire from tapping the HT side.
**broken link removed**
This doesn't help with the speed issue, but may be of some use.
Regards.
 
I was thinking these coils are sealed units with no access to the LT side. But there is a lug on the back side of the coil used to ground out the system to stop the engine and there is a good chance that is on the LT side.

$(KGrHqFHJBUE9p8TD!fVBPhlNWcdrQ~~60_35.JPG?set_id=8800005007
 
Last edited:
A rectifier and a small value capacitor to the ground should be able to get and hold the maximum value for a while until you measure it.
 
If you just want to detect when the spark takes place an inductive pickup is a very reliable and trouble free method, professional time strobes used to use this method.
To measure the ignition voltage you could do it with a string of resistors as a divider into a op amp circuit, but it would need to be high impedance to avoid pulling down the ignition voltage, it would also need to be low capacitance to avoid changing the coils resonant freq which would give you misleading results as the voltage might well change with freq.
Maybe you could do something that detects the electric field, shouldnt be too hard with ht, maybe a pcb track carring the ht voltage with a parallel track going to a fet or something on those lines.
 
I was thinking these coils are sealed units with no access to the LT side. But there is a lug on the back side of the coil used to ground out the system to stop the engine and there is a good chance that is on the LT side.

That connection would work fine if you are planning on a system that would have originally connected to the points and condenser lead being circuit wise that is the same tap point on the coil.

However I am not so sure how accurate it would be fore a true plug voltage reference point.
 
I have worked on cars a bit and am familiar with inductive timing lights. The timing on these little engines are generally fixed unless you use offset keys or something.

A rectifier and a small value capacitor to the ground should be able to get and hold the maximum value for a while until you measure it.

This is something I can try. Sort of like a compression test but we are cranking up a capacitor. Simple if it can be made to work.

Any ideas on where to start with components? I need to look at diode switching speeds. Wonder if small signal diodes would work.

I do appreciate all the replies.
 
If the coil only has a kill terminal, then it maybe a CDI coil like this one of a brush cutter. Heres a circuit of one of a Ryobi I dissected.
 

Attachments

  • Brush cutter CDI 2.JPG
    Brush cutter CDI 2.JPG
    113 KB · Views: 470
  • Brush cutter CDI.JPG
    Brush cutter CDI.JPG
    109.1 KB · Views: 516
KISS I was thinking of going for the LT side and using a full bridge rectifier. That is if that lug is the LT side. I have a multi meter that does 1000V DC max/min that samples 5 time a second. With the cap setup it should work.

The HT side would be better because then I would be test everything but the spark plug. Will have to order the parts to do that.
 
Last edited:
You know the polarity, right? Then you can put a voltage divider to reduce the voltage to +5V or so, then single diode, then capacitor. The capacitor will charge through the diode during the peak. Then diode will prevent it from discharging until you measure it. The diode will also protect your ADC from negative voltages. You will have to figure out the diode drop in your calculations, but looks like you don't need high precision anyway.

The only bad thing that may happen is that the diode leakage current may discharge the capacitor faster than you can measure.
 
Heres an idea.
You could use another ignition coil as a step down transformer, connect another coil to the existimg one, ht to ht, then the primary will be a transformed down version of the ht to the 'plug.
So long as you know the turns ratio you can work out the voltage.
There might be some loading effect to the system, you'd have to try it and see.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top