Ignition Advance When Knock Control Active

Post Reply
dr_jones
Posts: 162
Joined: Wed Jun 08, 2016 2:24 pm

Ignition Advance When Knock Control Active

Post by dr_jones »

Can anyone clarify just how 'Ignition Advance When Knock Control Active'?

..Its not clear from the manual or the sCAL help whether:

1) Its applying the entire value at (x)RPM on top of ignBase and then relying on the knock control to step back to a 'safe' value
or
2) Whether its gradually applying advance each cycle up to the value specified in the 'Ignition Advance When Knock Control active' table at (x)RPM until knock is detected (if it indeed is) and then retarding.

I suspect it's (1) but as I'm just looking into configuring this for the first time it has some bearing on how much advance I dial into it...

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Ignition Advance When Knock Control Active

Post by RICE RACING »

It's 1)

I use it by setting the base ign table a few degrees back from where its still safe (away from knock in worst operating case) then I typically dial in up to 3 deg via this function as referenced against the rpm axis.

Reason for doing it this way is if the knock control is not active then its got an ign map with more margin in it for obvious reasons.
This is one Mazda running it, 3150mB MAP in 35 deg C ambient temp conditions, and its on knock limit most of the time > https://www.youtube.com/watch?v=u44_xU-rCkY
Anyway that is one of the strategies I use, though again you have to have all of the knock control set up spot on prior to using this.


p.s. It works really well (once you dial in all of the other settings) I use it extensively on rotary (and piston) engines which are 'hand grenades' and lots of other *experts* say knock control will not work, well it does on this ECU platform :)
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

dr_jones
Posts: 162
Joined: Wed Jun 08, 2016 2:24 pm

Re: Ignition Advance When Knock Control Active

Post by dr_jones »

Thanks for clarifying - certainly seems to be doing the trick in the video there...

Mark_r33
Posts: 26
Joined: Sat Jan 18, 2014 7:14 pm

Re: Ignition Advance When Knock Control Active

Post by Mark_r33 »

Have you turned (m)any into hand grenades during the initial setting up of your knock control strategy?

I've always heard that the rotary is very easily upset by any knock, so would be interesting to hear how you go about dialing the knock sensing in in a properly "safe" manner.

That Mazda looks pretty savage!

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Ignition Advance When Knock Control Active

Post by RICE RACING »

Mark_r33 wrote:Have you turned (m)any into hand grenades during the initial setting up of your knock control strategy?

I've always heard that the rotary is very easily upset by any knock, so would be interesting to hear how you go about dialing the knock sensing in in a properly "safe" manner.

That Mazda looks pretty savage!
Not specifically no (only one had a 'soft failure' read below)

There was an engine supplied by PPRE that used their brand of product that had a failure within 250km run in period, that one had run away knock that was not able to be controlled by the ECU and while it showed no signs of broken parts (no dented rotor, corner seals cracked etc) it did suffer apex seal distortion but that was at very low loads :shock: As usually the engine builder blamed everyone but himself, so my customer refitted the stock standard 13B-REW engine of 15 year old vintage genuine factory OEM engine, that one was run as proof of the knock control strategy and it passed over 30 minutes of full on-load (rape) testing, which was up to 2850mB and 8500rpm. Customer then rebuilt the PPRE supplied engine, but this time with decent components replacing the failed items and that is the one in the above video linked, 3150mB+ 9100rpm on tap pretty much all the time. Its bullet proof so far.

What I have found:
You do have to use the right knock sensors
All software settings need to be specific to the rotary engine (which are quite different to 4cy to 10cyl reciprocating ones I do)
I tested my own knock sensors with a peco scope to get the right frequency on an actual engine first
Then followed the basics given to me by various engineers at LR
There are some specific items that make it unique but ironically I find it actually easier on the rotary than on reciprocating engine as they are mechanically 'quiet' by comparison so once you have your knock thresholds set correctly to suit and all other things done (see Ryan's video on subject) then its not too hard, and you wont break engines working it all out either ;)

Here is one of the many set up worksheets I use, shows it working and I record every individual cycle of the engine too ;)
You do have to run things a certain way to have no knock on action of rev limiters and traction control, once done it all works perfectly, as proven in this car.
Its not usable without the TC by the way, 275 wide R888 tires, still have to limit boost in 1st gear = ~100kph at 9100rpm. 2nd gear use full boost but needs the TC to keep her going at the rate in that video !

RIGHT click and view image for full size I guess?

Image

Unlike the pretenders and 'fake news' rotary self professed experts with a one off dyno sheet 'claim' this car get the RAPE dungeon treatment daily!
Image

and
Image
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Ignition Advance When Knock Control Active

Post by RICE RACING »

RICE RACING wrote:
Fri Sep 15, 2017 12:29 am
Mark_r33 wrote:Have you turned (m)any into hand grenades during the initial setting up of your knock control strategy?

I've always heard that the rotary is very easily upset by any knock, so would be interesting to hear how you go about dialing the knock sensing in in a properly "safe" manner.

That Mazda looks pretty savage!
Not specifically no (only one had a 'soft failure' read below)

There was an engine supplied by PPRE that used their brand of product that had a failure within 250km run in period, that one had run away knock that was not able to be controlled by the ECU ....
^ That part needs some more detail which upon re reading. The knock was 'out of control' (due to component failure inside the supplied engine) the ECU however did control the following pre ignition event trigger and shut down that part of the engine on the following engine cycle, thus preventing any additional damage ;)

So turbo rotary engines do tolerate quite a lot of knock and pre ign as well.

Attached: This is what the run away knock looked like and the root cause of it, again all bare the sub par sealing parts distorted during run in period which cause the hot gasses to leak back into the preceding chamber were totally fine and thus reused. Had another case with the same user who tried another set of equally dodgy parts, lasted a bit longer but again same situation, gas leakage around this time a worn out leading edge causing sever gas leakage and resulting in equally high knock readings, and again the pre ignition event trigger saved any further engine damage.

p.s. Seen the same thing (gas leakage) happen in various piston engines which results in the same high knock readings which then lead to pre ign or super knock events. Shown up during leak down tests, and again thorough initial set up of the knock control system prevents any further damage happening to the engines. It's the best thing about the Syvecs that you really can make the engine live through what on inferior offerings would have resulted in the whole motor being a bin job.
Attachments
screenshot.2 (3).jpg
screenshot.2 (3).jpg (602.71 KiB) Viewed 3335 times
gas leak out of control knock.jpg
gas leak out of control knock.jpg (415.95 KiB) Viewed 3336 times
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

MReilly
Posts: 24
Joined: Fri Jan 19, 2018 7:05 pm

Re: Ignition Advance When Knock Control Active

Post by MReilly »

Thanks for the update!

In the picture of the rotor, is that something melted onto the rotor, or has the actual rotor melted a bit there? Bit dark to tell in the picture on my computer is all?

When you say that you tested your knock sensors on the engine, are these some frequency specific knock sensors such as some of the OEM's use, or do you mean you watched some wideband sensors through the scope to see what frequency knock occurred at raw, without the ecu?

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Ignition Advance When Knock Control Active

Post by RICE RACING »

Nothing melted, pic shows light passing through where there should be no light (part failed) so gap between seal and outer housing.
Knock sensor tested on engine with scope yes, works perfectly. It is so good I could never run any engine without it again, especially a rotary.
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

MReilly
Posts: 24
Joined: Fri Jan 19, 2018 7:05 pm

Re: Ignition Advance When Knock Control Active

Post by MReilly »

Do you induce light knock to verify the frequency on the scope?

Do you then run with that exact frequency, or a harmonic of it?

I found on the Vauxhall / Opel / Holden Z20LE* engines with Link G4+ ecu that the knock channel was very noisy at the "calculated" knock frequency (from online calculators). I dug into the standard ecu to find that they (Bosch) were watching the second harmonic of the frequency that I was using, which I tried in the Link with much greater clarity on knock events.

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Ignition Advance When Knock Control Active

Post by RICE RACING »

I only recommend the use of LR based items for last 5 years, which do work as per the advice they give directly on how to set it all up.

It's funny you mention that brand, cause its that piece of absolute certified garbage that detonated two of my own personal motors (known manufacturer admitted hardware fault and one software upgrade fault), fwiw not just me either :roll: but I was the only one at that time who proved externally that they are rubbish. To their credit they paid me $7500 but they initially wanted to do it with that value of product !, I told them they could keep it :lol:

Finally I just woke up, open the wallet and invested my money & time to learn how to use and understand actual professional motorsports ECU's Pectel > Bosch > Life Racing/Syvecs. After owning/using those I settled on LR/Syvecs and never ever looked back, if it can make a fragile Wankel detonation failure proof then its worth it's own weight in gold :D
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

Post Reply