ech

OK, I've had a problem with wavering boost forever, well at lest since I've installed my PFC.

All (I mean ALL) of the usual suspects have been tried - new WG and PC solenoids, checked and double checked connections, all hoses, etc. I've tried every boost level and just about every base duty combination possible. I can hear and feel the wavering as the car pulses on the changing boost levels. I finally got my datalogit connected, and look at this!!!



I'm thinking either there is an electrical problem, or there is some weird harmonic feedback. Has anyone ever seen this with their datalogit?

Any ideas? - the wavering happens 2-3 times a second, and it is pretty regular. I don't think a boost leak could be this periodic, and this constant over time (and I've replaced the boost hoses anyway, and pressure tested the IC).


After looking at my other logs, and logging all the other data, my next steps are to:

1) take out the WG and PC pills and see if that changes anything...
2) maybe try a different boost control.
3) I have the stock ECU, but I'm afraid to do a WOT run with it and my mods. I guess I could try another PFC to see if mine is defective ???

any thoughts or help would be appreciated.

Badog

Erik,

Did you have boost problems before the PFC?

I would check the pills (have you ever messed with the pilled lines?), check the pots (where the pilled lines connect to...that control the arms) and then work your way back to the vacuum nest from there. Check the connectors going to the WG and PC solenoid. Those wires get brittle too.

Here is a chart from a dyno run...I was using the Apexi AVC-R at the time.

[/B][/QUOTE]

ech

Badog,

Thanks for the chart and the suggestions.
Dave at KDR checked the lines and the wires. He thought the connections to the Solenoids was bad and rebuilt it. The solenoinds are new, the turbos and pots are new, and the lines to them are in great shape (and short).

The trubos are '99 and the pills in the lines are stock (maybe regular stock) and the PC pill may be reduntant with the oriface in the nipple boss. I drilled out the "pill" in the WG boss already from a prior problem.

And no, this wavering boost problem was new with the PFC. It took me quite some time to notice it though. You can't feal or hear - or even see - the problem on the PC - it's only the top end on the WG that you can detect.

BTW my boost settings were
.80 56
.70 64
and I've changed them up and down and all around without a difference.
----

Badog, w/ the AVC-R the PC signals are still hooked up, right? and the WG signals go to an unused solenoid? The big dip in the PC looks a lot like I see all the time, it doesn't seem to have a reason to be there.

ech

Oops,
read the wrong map:
My boost settings were
.85 66
.80 64

Badog

AVC-R: They were as you describe, solenoid-bottom lines went to the disconnected solenoids. Then, recently, I disconnected the solenoids completely and capped off everything other than the AVC-R lines. Now, I have the solenoids back in, stock connections, and the AVC-R disconnected. I'll post a pull I did a few minutes ago when I can dig up that log.

I am watching your posts on the Datalogit group too.

ech

I was asked to give it a go with:

.85 68
.85 64

The results are in:


I also tried it with:
.85 58
.80 62

and there was no appreciable difference.
Studying the logs there are places where the PFC seems to be reacting to boost limit, and there are other places where it seems to be doing the wrong thing. The constant is the regularity.

I believe in trying the easiest things first. I tried tuning boost, no go. Next will be an PFC ECU swap, then after that changing the pills, etc.

I welcome any other ideas, and will post my findings from this weekend.

ech

And here is a WOT in 3rd and 4th pull - where I was verrrrry slowly pulling away from a modified M3 (newest one). The car is defineteley still fun. I didn't even notice the wavering, but the datalogging tells the story.

Badog

Well, you have any idea what your sample rate is on the Datalogit?

Duty cycle refers to an "binary" operation either ON or OFF. Since you're getting boost and still able to blow away a M3, I wonder if you're seeing the WG and PG cycle?

Wondering what a PFC swap will show.

ech

My logs are sampling at about 33-36 Hz - so the drops are comming on at about 3 times a second. - I don't think we are seeing the duty cycle.

Here is another thread with graphs from a working system.

https://www.rx7club.com/forum/showth...threadid=81185

The other thing I didn't mention is that the wavering shows up on my dyno runs. Here's a set of runs at 10lbs, in a 100+ degree garage. We dialed in more fuel and as the room and car got hoter the wavering got really bad - so that the end looked like the beginning.

BTW, check out the low-end from the '99 turbos and having BOTH cats (both from Tri-Point) :-)

Badog

Yah, if it shows up on the dyno runs, it is not right! Can't be the duty cycle. The dyno run is freaky with all that wavering!

BTW....Here is the thread on my dyno (that goes with that data logged session)

How about this torque curve from STOCK turbos, STOCK block?




Keep us up to date on what you try next.

Mercury

I think I've been experencing the same thing... I play w/h maping the waste gate later this weekend.

ech

Today I met a local FD owner, Bruce, who graciously agreed to swap PFC boxes. I was careful to ground myself while swapping boxes and restored my maps before firing up the car - and then let the PFC re-learn idle.

Went for a data-logging run and here is the result:


The car is ecactly the same other than the PFC box - even the datalogit unit, the computer, everything. The boost settings are the same as the first graph posted. My harness has the correct wires clipped - so even if Bruce's box didn't come from SR, then it should behave the same .... but it didn't. It obviously worked a bit better.

So I think I have pretty good evidence that there is something wrong with My PFC ECU.

I put my ECU back in - again with caution and with the same maps restored - and the old boost patterns returned - along with the wavering.

If I had to bet on what was wrong - I'd say it is a bad capacitor. But I've heard that the boards on the PFC are surface mounted, so there is no way I can really test or fix anything. Time to try getting support from by vendor and Apexi.

I have had the unit for more than a year - but for the first 4 months I was running with the PFS boost guage - with a restrictor pill in the line - and didn't notice the problem for a long time - and when I did it was time for a new engine (not related to this at all).

With the new engine KDR replaced the vacuum lines, including the one to the boost guage - and the problem showed itself. Dave thought it might be a bad check valve, but I had to take the car back, so he didn't have time to fix it.

In two subsequent visits to KDR, and a ton of diagnosis (the Datalogit wasn't out yet, or at least I didn't have it) we still couldn't find the problem. That took a year and I or Dave:
- checked all the vacuum lines
- replaced a long vacuum line
- replaced the PC and WG solenoids w/ new
- rewired the harness for above
- replaced the CRV and CRV elbow hose w/ new
- installed a y-pipe and replaced all hose couplings and clamps (some twice).
- pressure tested the IC.
- Tried tuning boost in the comander
- finally used the datalogit and ECU swap - AHAH.

So thanks to everyone who helped, and gave my suggestions. Esp thanks to the Datalogit team - that tool and the on-line support are amazing! Also very much thanks to Bruce for the ECU test.

I'll post when I know what happens with the ECU.

-Erik

ech

Arg! SR Motorsports was little help. They did have a running special for an out-of-box unit, which I ordered, but I got _no_ deal over what anyone else would have gotten. I was hoping to get some value from having paid full price for parts when I could have saved $$s. Oh well. It _has_ been almost two years.

SOoooo........

I'm selling this unit, which will actually be perfect for anyone running another boost control system. I really want to use the PFC boost control with the stock sequential twins. If anyone is looking to go single turbo or parellel with another boost controller (Profec B, manual valve, etc.)

I do want to stress that other than this boost control issue the unit is working great- I've checked over all the other tracable variables, and it is working well. It even does an passible job with boost control - but not a perfect one - obviously.

If anyone is interested, PM me. I'll also post on the FS forum.

ech

Wow - that was quick. I seem to have a buyer.