KLR Part 2

[soxnail]

After a long hiatus I am reviving this old topic.
Since I have had no luck finding schematics, I am starting to take a KLR apart to figure out how to fix/modify it.
As 1st step here is the functional block diagram I created from what I know so far...please let me know if any part is incorrect.
Thanks Sox



Here is the process (for y'alls amusement ).

[soxnail]

So next :
- WOT occurs at 60 to 65 % of TPS position.
- DME has no information about knock.
- Knock sensor signal conditioning is a band pass peaking at 7.2 Khz.
- Knock sensor signal is sampled at trigger input time.
- Knock signal for 2 ignition events will start retarding ignition out from KLR.
- Ignition retard is 3 deg per event to the point where the KLR will totally stop ignition pulses for
multiple engine revolutions.
- Knock detection does not turn off the "make boost" PWM to the CV.
- All faults detected other than knock will put the KLR into limp mode.
- Limp mode produces about 5 psi of boost ( what it takes to open the WG).
- CV PWM is 80% full on and about 15% when in bypass.
- Excessive knock will turn on diagnostic led and retard ignition but will not put KLR into limp mode.

I have empirically tested the above, if incorrect let me know. I will continue to further take the KLR apart and post as I go.
Thanks Sox

[soxnail]

Some signal wave forms from the KLR:

Pic #1 is the timing of the trigger signal and the ignition I/O.



Pic #2 is the Knock sensor input and the knock detected out put from the KLR. Note the sensor out put is being sampled at the trigger time.



More to follow Sox

[DannoXYZ]

Awesome! I've always wondered what signals on those lines looked like.

[soxnail]

Ongoing with KLR signals and timing:
The pic below shows the relationship between the TPS, WOT and the CV control pulses.
Looks a little cluttered, but the gist is:
At 57% TPS the CV is told to start making boost.
At 60% TPS the WOT is asserted to the DME.
The second pulse on the WOT line is just A/D converter threshold point as it is a low res (8 bit converter) and can glitch at the cross over point.
More to follow
Sox

[soxnail]

This is a sequence of boost control:



WOT is followed by CV duty cycle starting and going to 80% plus.
Boost pressure starts to go up .
CV duty cycle starts to fall.

This is a sequence of over boost protection:



Same as first pic except the boost is high enough to trigger protection and stop the CV pulses.

More to follow
Sox

[richardbowen]

Awesome!

buy auto parts

[soxnail]

So as far as I can tell this may answer the ongoing question (which may have been answered while I was snoozing)
Putting in a manual boost control will effect the KLR's ability to protect the engine in case of any events that need protection, but not for knock. So the knock protection is retained but the rest are lost.
If I miss read some thing please correct as usual.
Thanks Sox

[soxnail]

Originally Posted by soxnail

So next :
- WOT occurs at 60 to 65 % of TPS position.
- DME has no information about knock.
- Knock sensor signal conditioning is a band pass peaking at 7.2 Khz.
- Knock sensor signal is sampled at trigger input time.
- Knock signal for 2 ignition events will start retarding ignition out from KLR.
- Ignition retard is 3 deg per event to the point where the KLR will totally stop ignition pulses for
multiple engine revolutions.
- Knock detection does not turn off the "make boost" PWM to the CV.
- All faults detected other than knock will put the KLR into limp mode.
- Limp mode produces about 5 psi of boost ( what it takes to open the WG).
- CV PWM is 80% full on and about 15% when in bypass.
- Excessive knock will turn on diagnostic led and retard ignition but will not put KLR into limp mode.

I have empirically tested the above, if incorrect let me know. I will continue to further take the KLR apart and post as I go.
Thanks Sox

"- Knock sensor signal conditioning is a band pass peaking at 7.2 Khz. "
THIS INFO IS OFF, IT IS 6.52 KHZ
Sorry about that
Sox

[kdjones2000]

Sox, manual boost controller still allows DME to pull ignition timing, but as you said cannot pull boost.

Only standalones with knock modules can pull boost.

[soxnail]

Today's nugget,
After extensive testing (on my 1 KLR unit )
KLR cuts boost off at 15.2 or 15.3 PSI and turns on the high boost warning.

I still find it hard to believe, but the engine can be knocking itself to death but the KLR is not programed to cut boost ???
Sox

Edit: KLR has a minimum of 1800 rpm floor before it turns on the boost. If the rpm is below 1800 the tps being turned to WOT has no effect on boost production.
Sox

[Mike Goebel]

Wait, you said the KLR cuts off boost at 15.2-15.3 psi but then you said the KLR is not programed to cut boost? Me no understand.

On another note in a different thread you pointed me too, it showed that a bad knock sensor/harness can cause a low boost issue. Some boost can be achieved but not full boost. What causes the low boost situation to occur from a bad knock/harness?

Thanks
Mike G.

[soxnail]

The number I mentioned was for absolute boost cutoff. Nothing to do with knock.
The KLR will cutoff boost when it hits 15 + PSI.
The KLR will not cut boost if it detects knock.
Sorry about my confusing style of writing...(Engineers you know )
Sox
As to your second point...I am exploring it right now. IIRC he fixed his low boost issues by replacing the knock sensor.
The only way I see this happening is that with retarded ignition the engine just was not putting enough air thru the turbo ??
Your and my issue is down to WG letting go or turbo without enough oomph.
I am in the process of running a couple of tests. Will up date.
Sox

[Mike Goebel]

Originally Posted by soxnail

The number I mentioned was for absolute boost cutoff. Nothing to do with knock.
The KLR will cutoff boost when it hits 15 + PSI.
Sox

Does the KLR open the CV to direct manifold pressure to the wastegate control line to reduce boost when the 15+ psi threshold is met? Or is there an ignition or fuel cut or....?

Thanks
Mike G.

[soxnail]

It cuts off the pulse train to the CV and thus all air bypasses the Turbo. It will also turn on the diagnostic led code 3:2. Also the car will be in "limp" mode.
KLR has no control over fuel...strictly a DME thing.
KLR can retard ignition with knock, to the point of actually cutting it off to stop knock.
Sox

[soxnail]

The knock sensing will set the diagnostic code of 2:2 with knock detected, but as far as I can tell the code only shows when rpm drops to 1500 and is erased when the throttle goes to idle.
Sox

[soxnail]

Sorry about info coming in drib and drabs....
CV is run at about 90% duty cycle till 7psi of boost, and drops off to 15% by 9psi.
Sox

[Mike Goebel]

Small dosses are easier to read.

Thanks
Mike G.

[soxnail]

Here is my take on the KLR and protection...feel free to correct as always.
If you use a manual boost control you loose the following:
-Over boost protection.
-Any fault that would put the car into limp mode. that's pretty well every one other than Knock.

You will retain the protection from knock.
Sox

[commdiver]

2 bar map sensor installed bye Bosch limits the klr to only reading 14.7psi once you go over that with modifications then more harm then good is done if you want to make more boost than 14.7 psi you are forced to go it alone with little engine protection unless you go to standalone thanks for the understanding of the way klr works