[info] Canada coming in loud and fast

[RichyBugs]

Hey guys, just a quick intro, I'm a Canadian as was obvious by the title but I came across RusEfi a couple days ago just as I was going to grab a standalone ECU; the ingenuity and the actual fact that this is being done makes me want to support this. A quick background on me, I have been into cars my whole life, car guy through and through, however have not had a chance to build anything until now due to Uni etc. I'm still a younger guy at 22 however have done plenty in relation to building other cars.

I will have plenty of projects going shortly and if I can successfully make this work I will be buying a couple of ECUs for projects, one will be a small 250cc motorcycle, a big 1100cc drag/cruise bike and most notably a Foxbody Mustang that we will be trying to get into the 8's with. As far as I know nobody has done any high HP builds on RusEFI. The fox will be in excess of 1000hp on a turbo Chev LS, with any luck the motor will be built next year and a RusEFI system will be running it (If anybody sees any limitation in the hardware in regards to this feat please let me know ASAP). I'm new to coding, soldering, and messing around with systems in general, I have experience with Carbs...that's it. but if you guys are interested in seeing this thing run feel free to lend a helping hand (trust me, it will be needed) technically or even on the car itself if you're in the Toronto area. Also, if you are in the Toronto area give me a shout if you need help or sandblasting.

At any rate, I will be asking many questions over the next year as I build things and stuff; I hope that it works well and that we can maybe even getting things like nitrous control, launch control via trans brake and all the racing things going.

[AndreyB]

Hello and welcome!

@ is also in Canada "only" four hours away from you?

Actually good question what's the highest HP on rusEfi so far but probably around 200hp? Really guessing here.

So what's the next step? Have you seen http://rusefi.com/wiki ?

[RichyBugs]

oh wow, so close (for Canada) lmao, well, the way that I have played with ecu's in the past the control systems are quite...simple. AFR, knock sensors, correction, and timing, the mechanical bits should stand up to the stresses of 30psi+ but the issue I have currently is whether or not RusEFI will ensure that there is no detonation. it is easy to control that preemptively using egt, iat, and the aforementioned items. The issue is, how responsive have others been able to get their setups, a v8 running a huge turbo around 8000rpm can be difficult to manage, if the crank sensor misses or if the hardware misses the motor can kaboom itself very quick. The wiring seems straight forward, I looked between the schematics and pin outs available and I shouldn't have issues with the wiring. My main issue will honestly be coding the thing and making sure that it has all the inputs and they are being read accurately; along with that, calibration of sensors (i see that some are added already). I may be getting a friend of my to solder the deal together as this is what he does, however, once everything is together and wired up I will definitely need help getting the software going.

[AndreyB]

the issue I have currently is whether or not RusEFI will ensure that there is no detonation

We have detonation-detection hardware on the board and we have some relevant code in the firmware, but I am not satisfied with test coverage.

Meaning someone (maybe myself later) needs to induce knock and confirm that it's detected fast and reliably.

[kb1gtt]

Welcome along, and feel free to ask questions. Even better additions to the wiki are helpful

For high power applications, I understand knock is worthless. There is topically so much noise form your valves, combustion, cam lobes, ect that you can't distinguish knock vs normal noises. OEM engines can do it because they are quite. Even with a quite engine, you still need windowing and other such tricks to do good knock detection.

EGT's are a good start. Typically ion sense is what people dream of for detonation detection. I do not know how you would get ion sense though.

I do not know of any physical issues. You may need some interface boards if you are doing things like LowZ injectors. But for the most part I believe Frankenso has the IO you are looking for. What I'm not sure about is the software maturity. I think you would be the first to use several bits of code, those bits of code could cause some problems. I think turbo control is a lesser tested chunk of code right now. So there is some risk in bits of code like that, but I think that's just because @russian doesn't have hardware to work with. I believe @russian can write the code that does not exist, and fix bugs in code that may already exist. So I think it's possible, but you may want to start small and work you way to something bigger to avoid running into bugs.

Again welcome along and pleasure to meet you.

[RichyBugs]

I'll let you guys in on a little secret, the factory 5.3 with a cam, adjusted ring gap, and springs/retainers will last over 1300hp

So even if the motor blows up....I can get another for 300 bucks.

I wonder, would it be possible to also play with things like launch control and traction control based on abs sensors (which would also double as wheelie control) lmao

[kb1gtt]

Sounds like a good plan using the right combination of motor to keep the costs down.

From a physical standpoint I see no issues with things like traction control, ect. It would need some add on PCB's to get the wheel speed into some MCU digital IO signals. As well I expect we have more than enough clock cycles to get the tasks done. The key problem would be software generation, then software validation. That is a very significant task.

At this point in time, the features you mention basically do not exist, but could exist with some development cycle.

[RichyBugs]

I would be more than happy to provide a mechanical test bed capable of +1000hp. Start small at maybe ~500hp and slowly jack up power as software comes long? My main issue is that I have never had to interpret timing graphs etc (I'm sure I'll learn) but this project will enable to software to grown exponentially. As for the tc, what if we used Abs sensors on the front and rear wheels and cross referenced the wheel speeds of the axles? That would also double as wheelie control; as the front wheels come off the ground the motor cuts power ever so slivhtly.

[monettes]

Sorry to reply so late!

If you have cheap engines to test detonation with, I can put the test engine on my engine Dyno. My Engine dyno season is over now and I can play with stuff until February (dyno seasons starts in February and ends in July).

I will also start my own project with a gen 3 Hemi but this is an expensive engine to test detonation with! lol

Let me know.

Stephan Monette
Klodus Performance.

[AndreyB]

Welcome back and let's move things forward

[monettes]

I just need to finish setting up the electric motor on the water pump to run the 5.7L Hemi on the dyno. Got 2 more dynos this week! Sorry!

[AndreyB]

Looking forward

[RichyBugs]

I'm so late! Thing got bogged down, but finally are moving along once again, the car should be rolling later this week; currently welding together trans crossmember and modifying the k member. As far as te ecu goes I might be picking one up soon, yeah man, I should have an engine to play with soon! One got water in it and needs to be torn apart anyways. For now it seems that we will need to run the car based on afr's, no other sensors will be on the car for now, so as long as we have a strong spark and proper afr it should make power.

[RichyBugs]

[quote=monettes post_id=26611 time=1499050174 user_id=668]
Sorry to reply so late!

If you have cheap engines to test detonation with, I can put the test engine on my engine Dyno. My Engine dyno season is over now and I can play with stuff until February (dyno seasons starts in February and ends in July).

I will also start my own project with a gen 3 Hemi but this is an expensive engine to test detonation with! lol

Let me know.

Stephan Monette
Klodus Performance.
[/quote]

yeah man, for sure, should have something soon. Where are you located exactly?