CDI ignition and MSnS

All Megasquirt 1 support questions. See also the Documentation

Moderators: jsmcortina, muythaibxr

dart73
Helpful MS/Extra'er
Posts: 124
Joined: Tue Sep 21, 2004 11:12 am
Location: Brazil

CDI ignition and MSnS

Post by dart73 »

Have someone used an CDI ignition with MSnS extra??

I've bought a HEI-8 module to use with my dizzy... but... in a V8 at high RPMs I think it should help....

Have someone done you own CDI??? I've found a article about it...

any DIY CDI's???

Thanks,
Fernando
MOPAR or NO CAR!!
1973 Brasilian Dart
megasquirted
ami8break
Experienced MS/Extra'er
Posts: 203
Joined: Sun May 02, 2004 11:45 pm
Location: Austria, Graz
Contact:

Post by ami8break »

Hello Fernando,
I see no one replied. :(

I found this (and many others) CDIs too when I looked for a DIY CDI.
But I found none which is really easy to do (few and common parts and well documented).



My favourite type is "Sparkrite". This is a simple DIY CDI from the 70ties. I found a guy in Malaysia who is using this circuit sucessfully for 30 years. He was so friendly to scan the circuit but he doesn't know the numbers of windings.

Can anyone help to find out/calculate/design the coil to finish the circuit? (I'm not an EE.)

It would be fantastic if I can use this ~450VDC source for measuring ion sense current (~10mA) too. (Refer to "ISP" labeleed topics in Hitech subforum)

Thanks
»Horst
aerowerks
Helpful MS/Extra'er
Posts: 48
Joined: Tue Jun 29, 2004 4:18 pm

Post by aerowerks »

All of these systems are relatively standard setups for CDI. The newest stuff uses switching power supply (like in computers) to get some 600 Volts to the coils. Remember energy stored in a capacitor is 1/2*C*V^2, by increasing the capacitor voltage (400 to 600 Volts) you will get 2.25 times the energy to the spark (neglecting wire and other loses). More energy to the spark means better and more reliable charge ignition. The switching power supply is much more compact and much more efficient than a step up transformer system like in the schematics above.
ami8break
Experienced MS/Extra'er
Posts: 203
Joined: Sun May 02, 2004 11:45 pm
Location: Austria, Graz
Contact:

Post by ami8break »

Hi aerowerks,
I'd be very satisfied if I can run a simple 'standard CDI setup', but I really wouldn't complain if you can show me a more modern circuit which I can solder at home. :)

I've read some patents/papers who point out to minimize ignition energy for each case/condition. IIRC the reason was to lower manufacture and component costs and extend spark plug replacement intervalls - but maybe I'm wrong.
In lean engines or other advanced technologies more energy is necessary to get a reliable brurning, my aim is to get around stoechiometric mixture so typical induced energy would be enough.

»Horst
aerowerks
Helpful MS/Extra'er
Posts: 48
Joined: Tue Jun 29, 2004 4:18 pm

Post by aerowerks »

You are correct, lower energy supplied to the spark plug will extend the replacement period and high energy circuits are used for lean burn engines, also nitrous, alcohol, and nitromethane require much greater ignition energy than gasoline. My reasoning for using very high energy ignition systems lies in exactly that, lean burn engines. By supplying greater energy to the lean mixture, the high energy ignition will cause the mixture to ignite and burn faster and more completely so you can avoid the dreaded long burn into the exhaust portion of the stroke (burned valves and excessive exhaust temperatures). Basically this boils down to running a lean engine on cruise = better fuel economy. That is how Toyota and Honda get such incredible fuel economy. Since the spark on CDI systems is much shorter than on Inductive systems, the plug doesn't get as much "wear", you can also run a colder spark plug and extend the service time that way (also helps with preignition and detonation).
I'm working on a high energy ignition system using the switching power supply in my spare time. It will have separate trigger for each coil so you can wire it directly to MS'n'S and uses a Linear Technology IC. Since I'm a full time grad student, spare time is a luxury. When I finish the design I'll post it so people can try it.

Rick
Last edited by aerowerks on Tue May 10, 2005 1:54 pm, edited 1 time in total.
Astrona
Helpful MS/Extra'er
Posts: 98
Joined: Mon Sep 27, 2004 1:37 pm
Location: Estonia
Contact:

Post by Astrona »

I have a 220V DC2AC 300W Adapter in car.
And i'm also NOT an EE.

But maybe we could simplyfy (dont know how) CDI with these adapters. They cost only 40 to 75$ and can be used to charge laptop anyway.
S13 CA18DET GT2876R
ami8break
Experienced MS/Extra'er
Posts: 203
Joined: Sun May 02, 2004 11:45 pm
Location: Austria, Graz
Contact:

Post by ami8break »

WARNING!
James told me that shematic sparkrite.gif (Posted: Mon 04 Apr, 2005) has an error, not easy to see on the scanned documentation, but on the layout for the PCB.
Please take a look to sparkrite_251-error_113.gif (see the red mark).
»Horst
Astrona
Helpful MS/Extra'er
Posts: 98
Joined: Mon Sep 27, 2004 1:37 pm
Location: Estonia
Contact:

Post by Astrona »

Astrona wrote:I have a 220V DC2AC 300W Adapter in car.
And i'm also NOT an EE.

But maybe we could simplyfy (dont know how) CDI with these adapters. They cost only 40 to 75$ and can be used to charge laptop anyway.
And please explain, what is bad with that idea?:)
S13 CA18DET GT2876R
aerowerks
Helpful MS/Extra'er
Posts: 48
Joined: Tue Jun 29, 2004 4:18 pm

Post by aerowerks »

Astrona

There is nothing wrong with using the DC2AC converter for the supply side of CDI, the voltage is little low, but it could be used. I don't know if it is a step-up transformer supply or boost-flyback switching supply. Switching power supplies have efficiencies on the order of 80% and higher, whereas step-up transformer supplies struggle to get 40%, so 60% of energy going in turns into heat.

Rick
ami8break
Experienced MS/Extra'er
Posts: 203
Joined: Sun May 02, 2004 11:45 pm
Location: Austria, Graz
Contact:

Post by ami8break »

Hi Astrona,
maybe with a 220VAC resource we could 'copy' a more simple magneto CDI (like used in older bikes and most scooters).
Battery CDI requires ~400VDC - don't know why.
»Horst
aerowerks
Helpful MS/Extra'er
Posts: 48
Joined: Tue Jun 29, 2004 4:18 pm

Post by aerowerks »

Horst

Its not that a battery CDI requires 400Volts, Its just that it seems to be a convenient number and one class of capacitors has that as its maximum voltage. My XR600's CDI voltage varies from roughly 150V at idle to over 400V at redline. Its a matter of how much energy you want to throw into the spark. It would be the same with your suggested "magneto" setup, low voltage at cranking and high at full revs.

Rick
ami8break
Experienced MS/Extra'er
Posts: 203
Joined: Sun May 02, 2004 11:45 pm
Location: Austria, Graz
Contact:

Post by ami8break »

Hello Rick,
yes my message was easy to misundterstand or I was wrong.
I tried to say that battery CDI has/needs a DC recource and magneto CDI is powered by an AC resource - at least I belive that.
Thanks for giving an explanation for the 400V potential - good to know that there is no myth for the ever read number of 400V. :)
»Horst
slow_hemi6
Super MS/Extra'er
Posts: 4122
Joined: Fri May 07, 2004 3:33 am
Location: Australia

Post by slow_hemi6 »

Well for a good cdi that adresses many of traditional cdi's problems you might condider this http://www.dse.com.au/cgi-bin/dse.store ... View/K3307
This design will start a car with an almost flat battery unlike many cdi and has multi or single spark output.
Find the Manuals up top under Quick links: Manuals. :RTFM:
Cheers Luke
ami8break
Experienced MS/Extra'er
Posts: 203
Joined: Sun May 02, 2004 11:45 pm
Location: Austria, Graz
Contact:

Post by ami8break »

Thanks slow_hemi6,
this is exactly the CDI from the "Silicon Chip" article (Cap-dis-ignit-CDI.pdf) added by dart73 at the beginning of this topic.
My 2cyl engine has no dizzy --> I'd need 2 seperate CDI units otherwise I wouldn't get rid of the wasted spark system which (maybe) causes additional problems for ISP.
Even one unit would be too expensive (including shipping and money transfer costs) for my budget.

I *know* that a simple circuit is enough for my purposes, but have no idea how to *design*/make a working 400V coil. Could I add the coil of Cap-dis-ignit-CDI.pdf article to the sparkrite_251-error_113.gif circuit?

»Horst
aerowerks
Helpful MS/Extra'er
Posts: 48
Joined: Tue Jun 29, 2004 4:18 pm

Post by aerowerks »

Horst

Winding the transformer is a relatively simple procedure, but requires a lengthy explanation. The ratio of secondary to primary number of turns will give you an approximate multiplication factor minus about 10% for losses, i.e. the Cap-dis-ignit has 13 turns primary, 360 turns secondary. This makes for a multiplication ratio of 27.7, thus 12V going in will give you 332 volts, minus 10% results in about 300 volts on the capacitor. To wind the transformer for the Sparkrite, do the same procedure as described for the Cap-dis-ign and use the same core and wire. Wind the secondary with 360-400 turns, the two primary 13 turn windings and add another winding of about 10 turns (same wire as the 13 turn to keep it simple) for the driver feedback (the top coil on the transformer in the schematic). You don't necessarily have to build two indpendent CDI boxes, just copy the firing circuit (inside the green outline) and connect them at the blue dots. You can repeat this for each firing "channel" you need. (2 for twin or waster spark four cylinder, 3 for wasted spark even fire six, 4 for coil on plug four or wasted spark V8 )
Another way is to find a comercial transformer with 220-240Volt primary, center tapped 12Volt 5Amp secondary and another 6Volt secondary. Use the 12Volt for the driver side, 6Volt for the feedback, and the 220-240 for the high voltage side (which will now be twice the rated voltage so make adjustments to the rest of the circuit so it can handle the increased voltage)

Rick
aerowerks
Helpful MS/Extra'er
Posts: 48
Joined: Tue Jun 29, 2004 4:18 pm

Post by aerowerks »

Oops I forgot about the other message

The "magneto" setup on scooters and other bikes relies for the high voltage necessary by having a coil on the generator that produces that voltage. In simple terms they disregard the rest of the Sparkrite circuit and use what is inside the green outline. The transformer is then replaced by the high voltage coil on the generator.

Rick
baldur
Experienced MS/Extra'er
Posts: 256
Joined: Sun May 02, 2004 6:39 pm
Location: Reykjavik, Iceland
Contact:

Post by baldur »

Of course making a multi channel CDI box should be pretty simple. Think about it this way: the number of sparks fired is the same as if you are firing a single coil and using a distributor, so the same step up converter could be used with no ill side effects. However there is one thing to look into, with 4 channels and 4 capacitors the startup current could be very high.
Baldur Gislason
baldur
Experienced MS/Extra'er
Posts: 256
Joined: Sun May 02, 2004 6:39 pm
Location: Reykjavik, Iceland
Contact:

Post by baldur »

I know that if I was to make a system like this from scratch I'd probably use one of the 30 or so PC power supplies I have around, that has a pretty good ferrite core transformer good for 200W. And feeding 13V into the 5V windings should give about 400V on the old primary. I have made various DC-DC converters using surplus PC power supplies but I have never dared measure the output of the old primary, but it has to be pretty high.
Baldur Gislason
slow_hemi6
Super MS/Extra'er
Posts: 4122
Joined: Fri May 07, 2004 3:33 am
Location: Australia

Post by slow_hemi6 »

One problem with the sparkrite is that it is very noisey oscilating nicely to give off rf. The silcon chip HV dc supply runs at 20khz to get outside this.
What if you went with the silicon chip supply then through two parallell legs with series 1n4936 diodes to 2 sparkrite output sections? I don't have the parts to try it but it would make for an interesting afternoon at the bench.
Find the Manuals up top under Quick links: Manuals. :RTFM:
Cheers Luke
aerowerks
Helpful MS/Extra'er
Posts: 48
Joined: Tue Jun 29, 2004 4:18 pm

Post by aerowerks »

Yes, the startup current would be quite high, but since it is a transitional phenomenon lasting few miliseconds, and the power output of the HV is actually limited by the resistance of the secondary winding on the transformer it should not pose any problems.
The switching power supplies in computers run somewhere between 20kHz and 2MHz. Proper decoupling capacitors and chokes on the inputs and power lines will be necessary to eliminate any feedback into MS'n'S and the whole system will have to be housed in a shielded (preferably steel) box.
The whole point of the system is that you will be able to connect it to MS'n'S directly and will not need the dwell option. I wanted to use this on my V4 bike which fires every 90degrees and revs to 15000 RPM (4 ms between sparks). This way I could set it up as a wasted spark "V8" and not worry about getting second trigger or running out of spark juice at redline.

Rick
Post Reply