Author Topic: Catching car's ignition signal  (Read 6824 times)

0 Members and 1 Guest are viewing this topic.

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #25 on: August 05, 2020, 01:15:41 pm »
@joeqsmith - Like everything in life, having anything fixed is bad. Usually, you monitor the gear position and let it cut for as long as needed. So it differs from gear to gear. Also, retarding the ignition, rather than cutting it, is much smoother and effective.
 

Online mikerj

  • Super Contributor
  • ***
  • Posts: 3383
  • Country: gb
Re: Catching car's ignition signal
« Reply #26 on: August 05, 2020, 01:34:59 pm »
Fuel cut on the other hand is very slow and not really useful in my case

Why would fuel cut be any slower than ignition cut?  With either one a cylinder could be cut for a single firing cycle if needed.
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Catching car's ignition signal
« Reply #27 on: August 05, 2020, 02:24:37 pm »
@joeqsmith - Like everything in life, having anything fixed is bad. Usually, you monitor the gear position and let it cut for as long as needed. So it differs from gear to gear. Also, retarding the ignition, rather than cutting it, is much smoother and effective.
There are many many things in life that I like constant but I thought you were working on an  ignition kill and not your philosophy of life.   

On the older bike, the ignition timing, boost, shift points and kill times are all fixed.  It's KISS and reliable.  The gains from the shift down track would be small compared to say adding another PSI of boost or just going on a diet.   

I assume you have data showing the effects of the shift using various techniques.   That would be very interesting to see.  No need to disclose anything proprietary about what you are doing but rather just show the end result. 

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #28 on: August 05, 2020, 02:47:52 pm »
Quote from: mikerj
Why would fuel cut be any slower than ignition cut?  With either one a cylinder could be cut for a single firing cycle if needed.

On full throttle, the amount of fuel going into the intake is huge, basically, you could fire all injectors at once all the time and it would not make a difference (would just be more fuel consuming). It takes some time for the air and fuel to mix so cutting 1 or 2 cycles of the injector, does not make a a lot of difference in reality. This is needed when shifting gear because the reaction needs to be instant and the longer you unload the engine, the harder it will shift. Ignition on the other hand is what drivers the car so every interruption in that, disturbs the car, which is what is needed.
 

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #29 on: August 05, 2020, 02:57:27 pm »
Quote from: joeqsmith
There are many many things in life that I like constant but I thought you were working on an  ignition kill and not your philosophy of life.   
LOL :)

Quote from: joeqsmith
On the older bike, the ignition timing, boost, shift points and kill times are all fixed.  It's KISS and reliable.  The gains from the shift down track would be small compared to say adding another PSI of boost or just going on a diet.   

I assume you have data showing the effects of the shift using various techniques.   That would be very interesting to see.  No need to disclose anything proprietary about what you are doing but rather just show the end result.

Unfortunately, I cant show you the graphs as this is the knowledge we've gathered in the last 15 years in motorsport industry.

It's not just about the shift speed gain - the duration of the "cut" (and the cutting strategies used) determine other factors - stability of the bike/car in tricky conditions, smoothness of the shift (engine likes smoothness and so does the gearbox and all other components). If you have 24 hour race rough fixed shifting, you'll probably end up failing the race.
If you're too long on the cut, this brings all kind of troubles, so does cutting too fast. Each gear is different - let say you start loosing a gear in a race. Once this starts happening, with fixed timing, you'll just destroy everything and not finish. Cutting the engine (with combination of rev matching) helps on downshifting as well, cutting too long will drop the rpm and make it rough as hell, possibly loosing the grip. This can be a huge problem in the rain. Just to name a few.
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Catching car's ignition signal
« Reply #30 on: August 05, 2020, 04:41:42 pm »
Quote from: joeqsmith
On the older bike, the ignition timing, boost, shift points and kill times are all fixed.  It's KISS and reliable.  The gains from the shift down track would be small compared to say adding another PSI of boost or just going on a diet.   

I assume you have data showing the effects of the shift using various techniques.   That would be very interesting to see.  No need to disclose anything proprietary about what you are doing but rather just show the end result.

Unfortunately, I cant show you the graphs as this is the knowledge we've gathered in the last 15 years in motorsport industry.

It's not just about the shift speed gain - the duration of the "cut" (and the cutting strategies used) determine other factors - stability of the bike/car in tricky conditions, smoothness of the shift (engine likes smoothness and so does the gearbox and all other components). If you have 24 hour race rough fixed shifting, you'll probably end up failing the race.
If you're too long on the cut, this brings all kind of troubles, so does cutting too fast. Each gear is different - let say you start loosing a gear in a race. Once this starts happening, with fixed timing, you'll just destroy everything and not finish. Cutting the engine (with combination of rev matching) helps on downshifting as well, cutting too long will drop the rpm and make it rough as hell, possibly loosing the grip. This can be a huge problem in the rain. Just to name a few.

I collect a very little data compared to say an F1 team but it would be trivial to show you what I have.   You've never mentioned what the application was.  Maybe you are not allowed to log the data.   I wondered that with the control.  I've seen in some racing were the do not allow for any type of electronics control.  Don't ask me how that would even work today.   

My hobby is a fair bit different.  You have 24hours of racing, I have a few seconds.   You have to be concerned with downshifting.  The only time I think about that is going from third back to first after warming the tire.  The air cylinder can only upshift.  Any downshifting is done manually.  I've damaged transmissions before I switched to using the Robinsons.  Because of the auto, as I mentioned they can be damaged by backing off the throttle.     In one of your 24 hour races, you would have traveled more miles that I would put on my bikes for their entire life.   

Interesting to hear more about your team and what they are doing.  Do you have a website?
« Last Edit: August 06, 2020, 12:40:26 pm by joeqsmith »
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Catching car's ignition signal
« Reply #31 on: August 06, 2020, 03:14:47 am »
Quote from: mikerj
Why would fuel cut be any slower than ignition cut?  With either one a cylinder could be cut for a single firing cycle if needed.

On full throttle, the amount of fuel going into the intake is huge, basically, you could fire all injectors at once all the time and it would not make a difference (would just be more fuel consuming). It takes some time for the air and fuel to mix so cutting 1 or 2 cycles of the injector, does not make a a lot of difference in reality. This is needed when shifting gear because the reaction needs to be instant and the longer you unload the engine, the harder it will shift. Ignition on the other hand is what drivers the car so every interruption in that, disturbs the car, which is what is needed.

What about diesels and HCCI?

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #32 on: August 06, 2020, 05:37:53 am »
Diesels have very precise fuel timing so it's different of course. In that case, fuel cut is the way to go. But I'm doing this for a petrol engines :)
 

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #33 on: August 06, 2020, 07:31:33 am »
Quote from: joeqsmith
Interesting to hear more about your team and what they are doing.  Do you have a website?

We do not have a team (although it started with me racing and doing stuff for myself).
We manufacture high performance paddle shifting (currently only pneumatic actuation) for various teams and automotive sports. Including 24 hour races, drag racing, hill climb like pike's peak etc.

Our web page is http://www.mme-motorsport.com
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Catching car's ignition signal
« Reply #34 on: August 06, 2020, 12:37:45 pm »
Looks like you make some very nice products. 

I wonder if you do any sort of HIL testing?   This like shows a bit of my simulator.  It's basically all digital and can provide timing data for the injectors and ignition.   I can't use it to tune an engine but for testing out various shift algorithms, it works great.   You can program things like clutch slip.   

https://youtu.be/q_89qoFMivg?list=PLZSS2ajxhiQBvWvqMVLdRQMjGofKpQUJr&t=2313

With this setup, both the injectors and ignition are modeled from standard parts.  The ECM had no problems with the fuel injectors but it wants something fairly close for the coils.  This engine doesn't use wasted spark like my older ones.  I was able to simulate all four coils with a common core and not have it fault out.  It was a bit of a surprise as they obviously couple but their firmware must ignore the signals except when the selected driver is active.     Obviously there's no need for any high voltage so I think I stayed with a 1:1 turns ratio.   There is a zener/FET to simulate the plug's breakdown.   

You may need to try several cars you plan to target and just see how they work.  Maybe there is a generic solution. 

« Last Edit: August 06, 2020, 12:55:50 pm by joeqsmith »
 

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #35 on: August 07, 2020, 07:04:31 am »
Nice setup you got there. The fuel system simulator - is that your own creation?

I usually deal with aftermarket ECUs, this is the 2nd or 3rd time that I'm trying to do anything "ignition like" on the standard ECU. I have engine simulator, that is able to generate wheel triggering and if you hook up original ECU, it does the cranking, firing up the coils etc. I cant use it in this case as this particular ecu has immo so it doesn't fire up and I wanted to make it simple. As I said, when ecu is operating with low power signals, I can handle that and I'll probably just make an external box that will catch high power signals and convert them into low power (and vice versa).

I'm going to try few other ideas from this topic next week and report if i'm able to get anything on the scope. Current measuring seems pretty good idea although catching correct dwell will probably be tricky in this case due to different firing ramp. What I dont want is to have this tweaking for every other car. Seemed very straight forward job from the distance :(

 

Online mikerj

  • Super Contributor
  • ***
  • Posts: 3383
  • Country: gb
Re: Catching car's ignition signal
« Reply #36 on: August 07, 2020, 08:08:24 am »
Quote from: mikerj
Why would fuel cut be any slower than ignition cut?  With either one a cylinder could be cut for a single firing cycle if needed.

On full throttle, the amount of fuel going into the intake is huge, basically, you could fire all injectors at once all the time and it would not make a difference (would just be more fuel consuming). It takes some time for the air and fuel to mix so cutting 1 or 2 cycles of the injector, does not make a a lot of difference in reality. This is needed when shifting gear because the reaction needs to be instant and the longer you unload the engine, the harder it will shift. Ignition on the other hand is what drivers the car so every interruption in that, disturbs the car, which is what is needed.

Great in theory, but popular motorcycle quickshifters such as the Power Commander manage to work perfectly well by cutting fuel.
 

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #37 on: August 07, 2020, 08:21:39 am »
Motorcycle has never been hard to shift at full throttle. They're very light, basically no load on the gearbox, low torque and very high rpm - you can shift that as long as you're able to disturb it a little. With cars, it's a lot different. Add a turbo with huge load and it will very different scenario. That's why, the solenoid shifters work well on bikes, but struggle with cars.
 

Online mikerj

  • Super Contributor
  • ***
  • Posts: 3383
  • Country: gb
Re: Catching car's ignition signal
« Reply #38 on: August 07, 2020, 10:19:21 am »
Motorcycle has never been hard to shift at full throttle. They're very light, basically no load on the gearbox, low torque and very high rpm - you can shift that as long as you're able to disturb it a little. With cars, it's a lot different. Add a turbo with huge load and it will very different scenario. That's why, the solenoid shifters work well on bikes, but struggle with cars.

No load on the gearbox because the quickshifter cuts fuel to remove loading!

A friends rally car has a 2L engine making around 285bhp at ~9k RPM with a Sadev sequential gearbox which uses fuel cut for shifting as well, so car engines are not really different.  However with a monster turbo pushing large quantities of fuel into the cylinder I can understand how this could be a problem.
 

Offline cmcraesloTopic starter

  • Regular Contributor
  • *
  • Posts: 64
  • Country: 00
Re: Catching car's ignition signal
« Reply #39 on: August 07, 2020, 11:26:13 am »
@mikej - would love to see your friend's parameters for the flatshift. Does he only uses fuel or adds fuel cut to the ignition cut/retard combination (that would be my guess).
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf