Pelican Parts
Parts Catalog Accessories Catalog How To Articles Tech Forums
Call Pelican Parts at 888-280-7799
Shopping Cart Cart | Project List | Order Status | Help



Go Back   Pelican Parts Technical BBS > 1- Porsche Technical Forums > 911 / 930 Turbo & Super Charging Forum

Reply
 
LinkBack Thread Tools Rating: Thread Rating: 3 votes, 5.00 average.
Author
Thread Post New Thread    Reply
Registered User
 
Join Date: May 2015
Posts: 616
Thanks Jeff. I confirmed from the hose diagram in the manual this morning that I have it plumbed backward. Will change that this weekend. My understanding is the temperature switch in the top of the crankcase near (not in) the ToD grounds the power circuit to the switch when it is COLD. Ground is interrupted when its HOT. IOW the switch is powered/on and the vac line from TB to dizzy blocked when cold. LMK if I have that wrong.
Old 04-28-2017, 03:41 PM
  Pelican Parts Technical Article Directory    Reply With Quote #821 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
ok some observations from the data I have, which isn't that great unfortunately.

I have EGO control ON above 1,300 rpm, 35*c, and 50 kPa MAP. I've done a bunch of auto tuning.

I made a couple of pulls to redline in 1st and 2nd. This was on public streets and didn't get a chance to keep going to 3rd, but I don't think it matters much, the engine doesn't know or care what gear its in, just rpm and map.

At partial throttle low power, and especially in overrun, AFR doesn't track target that well. That doesn't really surprise me, and I don't really care.

The system doesn't react quickly to throttle blips either from idle or when downshifting. Again, not surprising and I don't care.

Where I really care is high RPM, high MAP, high load conditions. This is where we care. In this situation I found the FCIS to track target AFR very well. I decimal place rich or lean here or there, but pretty much spot on.

Here is a log with a start up, long warm up and idle, then there's a little low power, then a couple of pulls almost at the end.

https://www.dropbox.com/s/9pa4f0gxb9ph24e/2017-04-22_15.39.14.msl?dl=0

I'll get some more soon.
Old 04-28-2017, 03:48 PM
  Pelican Parts Technical Article Directory    Reply With Quote #822 (permalink)
Registered User
 
Jeff NJ's Avatar
 
Join Date: Nov 2013
Location: NW NJ
Posts: 1,094
Quote:
Originally Posted by flightlead404 View Post
Thanks Jeff. I confirmed from the hose diagram in the manual this morning that I have it plumbed backward. Will change that this weekend. My understanding is the temperature switch in the top of the crankcase near (not in) the ToD grounds the power circuit to the switch when it is COLD. Ground is interrupted when its HOT. IOW the switch is powered/on and the vac line from TB to dizzy blocked when cold. LMK if I have that wrong.
That's exactly correct and why plumbing it backwards lowers your rpm because it retards, rather than advances timing. Getting that circuit working and the vacuum going through that solenoid really helped my car run better at cold start.

I would fix that before you do more pulls because the way you have it plumbed, that vac port on the distributor is closed at all times once warm. Not sure if dangerous or not, but you may lose some boost retard at WOT??
__________________
1988 930 Venetian Blue
Old 04-28-2017, 04:15 PM
  Pelican Parts Technical Article Directory    Reply With Quote #823 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
Quote:
Originally Posted by Jeff NJ View Post
That's exactly correct and why plumbing it backwards lowers your rpm because it retards, rather than advances timing. Getting that circuit working and the vacuum going through that solenoid really helped my car run better at cold start.

I would fix that before you do more pulls because the way you have it plumbed, that vac port on the distributor is closed at all times once warm. Not sure if dangerous or not, but you may lose some boost retard at WOT??
Got it plumbed the correct way. Confirmed that on a cold start for a second or two (its 90f in the hangar today) the RPM is around 1,200 then suddenly cuts down to 950 where I've set it.

However, that has not changed the low RPM lumpy idle on hot some hot starts. Its a bit difficult to figure out the exact behavior, but I'm thinking this is the same problem I had with the eWUR installed, only perhaps to a lesser degree. That is, a very high control pressure right at starting, that gives me a very lean mixture, causing a lean stumble. Once a little of that pressure has bled off, the system can manage back to where its supposed to be.

So, the question is, what could cause that? Moving to a separate thread as its pretty clear to me now this isn't a FrankenCIS issue.

Last edited by flightlead404; 04-29-2017 at 02:03 PM..
Old 04-29-2017, 01:49 PM
  Pelican Parts Technical Article Directory    Reply With Quote #824 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
So I've been doing some more analysis on the log files I have. In general I'd say its working pretty well. I think I'm getting beyond my skill level however, and would be happy to share and learn from someone who understands better than I how to analyze these logs.

My current setup: Stock turbo WUR, all emissions removed, added an AFR, microsquirt with RPM pickup spliced in to the TD output of the CDI, power to the microsquirt spliced into the power supply to the CDI, and a MAP sensor teed into the vacuum line to the stock WUR. Disconnected the lambda valve from the under-seat CIS box and connected the MS. Very simple easy setup with no major surgery and could be installed or put back to stock in no time (literally, 5 minutes). My goal here was just to provide the extra fueling needed at high power with the upgraded turbo and exhaust (K27/29, RarlyL8 headers and hooligan, Tial .8 bar w/g) while still being largely stock and easy to put back.

It looks like I need to do some more auto tuning at higher power levels, and then some manual work, but its looking good. Just using the MS to add or subtract from what the stock WUR does seems to be working quite well. I have only been focused on high power, high MAP areas so far, because that's the main area of concern for engine damage. I'm tracking the AFR map pretty good, usually within a decimal place or two, except where there's a sudden change, like opening or closing the throttle. There's also a sudden change when the WUR switches over to boost control pressure. In these areas it takes the MS about half a second to react. In many of those areas I'm actually richer than the fuel map.

Steve, any way to manually configure the eWUR to more smoothly map the control pressure/MAP/temp directly?
Old 04-30-2017, 08:34 AM
  Pelican Parts Technical Article Directory    Reply With Quote #825 (permalink)
3 restos WIP = psycho
 
kenikh's Avatar
 
Join Date: Feb 2005
Location: North of Exit 17
Posts: 7,714
^ that's actually a very helpful description. I'm very tempted to try to this sooner than later.

Sent from my Nexus 6P using Tapatalk
__________________

- 1965 911
- 1969 911S
- 1980 911SC Targa
- 1979 930
Old 04-30-2017, 08:54 AM
  Pelican Parts Technical Article Directory    Reply With Quote #826 (permalink)
 
Now Available for Ordering:   101 Projects For Your BMW 3 Series 1982-2000  [more info]
Registered User
 
Join Date: May 2015
Posts: 616
Quote:
Originally Posted by kenikh View Post
^ that's actually a very helpful description. I'm very tempted to try to this sooner than later.

Sent from my Nexus 6P using Tapatalk
Happy to share my configuration, tunes, and install info with you. No sense you going through my pains :-)

I'll put together a single consolidated write up when I have the time.
Old 04-30-2017, 09:17 AM
  Pelican Parts Technical Article Directory    Reply With Quote #827 (permalink)
3 restos WIP = psycho
 
kenikh's Avatar
 
Join Date: Feb 2005
Location: North of Exit 17
Posts: 7,714
Quote:
Originally Posted by flightlead404 View Post
Happy to share my configuration, tunes, and install info with you. No sense you going through my pains :-)

I'll put together a single consolidated write up when I have the time.
Greatly appreciated! I'll be in touch. I already have a legacy EFI system (used for ignition only) in place so am working on building an adapter harness for the MicroSquirt.

Sent from my Nexus 6P using Tapatalk
__________________

- 1965 911
- 1969 911S
- 1980 911SC Targa
- 1979 930
Old 04-30-2017, 10:05 AM
  Pelican Parts Technical Article Directory    Reply With Quote #828 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
Quote:
Originally Posted by kenikh View Post
Greatly appreciated! I'll be in touch. I already have a legacy EFI system (used for ignition only) in place so am working on building an adapter harness for the MicroSquirt.

Sent from my Nexus 6P using Tapatalk
Interesting. I'm probably going the other way. Once I've finished this, I'm thinking I'll add a crank and cam sensor and let FCIS control my ignition as well. Then cams....then...
Old 04-30-2017, 11:44 AM
  Pelican Parts Technical Article Directory    Reply With Quote #829 (permalink)
3 restos WIP = psycho
 
kenikh's Avatar
 
Join Date: Feb 2005
Location: North of Exit 17
Posts: 7,714
Quote:
Originally Posted by flightlead404 View Post
Interesting. I'm probably going the other way. Once I've finished this, I'm thinking I'll add a crank and cam sensor and let FCIS control my ignition as well. Then cams....then...
We're in the same page. I'll maintain electronic ignition control.

Sent from my Nexus 6P using Tapatalk
__________________

- 1965 911
- 1969 911S
- 1980 911SC Targa
- 1979 930
Old 04-30-2017, 12:05 PM
  Pelican Parts Technical Article Directory    Reply With Quote #830 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
Quote:
Originally Posted by kenikh View Post
We're in the same page. I'll maintain electronic ignition control.

Sent from my Nexus 6P using Tapatalk
Since you have one, it makes sense to keep it. I am going to have the advantage of already having a "brain" in the FCIS, and a single place to program ignition and fuel.

I *think* I need a replacement crank pulley with a sensor on it, and some sort of cam sensor, which I think is something in place of the dizzy, or on the scavenge pump housing, plus some coils and the right plugs.
Old 04-30-2017, 01:47 PM
  Pelican Parts Technical Article Directory    Reply With Quote #831 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
OK Steve, looks like we're back in your court with this low RPM stumble on start up. I can pretty much reproduce this at will.

I just did 5 start tests.

For tests one and two, a cold start, and a slightly warmer start. The only change from my prior data was to swap the connector for the lambda valve from the Microsquirt to the original CIS connector for these 1st two tests. Lambda duty cycle controlled by CIS, I assume at 50% give or take. There was no initial low RPM stumble on either start.

3rd test I simply reconnected the FCIS to the lambda valve and confirmed that the low RPM stumble came back as before.

For the 4th test, with the FCIS still connected to the lambda valve, but disconnected power to the AFR. This test did cause the initial low RPM stumble for 15 seconds as before. This eliminates any possibility of the AFR calibration sequence screwing things up.

5th test was the same as the third, back to all FCIS connected and confirmed the stumble remains.

The low rpm stumble period seems to last anywhere from 13 to 24 seconds in my observations. It may be shorter the hotter the engine. The low rpm stumble ends exactly where the lambda frequency duty cycle makes a sudden drop of about 2%.

Can you think of any reason why the lambda valve duty cycle is changing? EGO control is off, and indeed it happens with the AFR even connected.
Old 04-30-2017, 04:50 PM
  Pelican Parts Technical Article Directory    Reply With Quote #832 (permalink)
3 restos WIP = psycho
 
kenikh's Avatar
 
Join Date: Feb 2005
Location: North of Exit 17
Posts: 7,714
Can you just go alpha-n on cold start?

Sent from my Nexus 6P using Tapatalk
__________________

- 1965 911
- 1969 911S
- 1980 911SC Targa
- 1979 930
Old 04-30-2017, 05:01 PM
  Pelican Parts Technical Article Directory    Reply With Quote #833 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
I'm using speed density. Doesn't alpha-n require a TPS?
Old 04-30-2017, 05:17 PM
  Pelican Parts Technical Article Directory    Reply With Quote #834 (permalink)
3 restos WIP = psycho
 
kenikh's Avatar
 
Join Date: Feb 2005
Location: North of Exit 17
Posts: 7,714
Yes. I forgot you don't have one. I installed one on my car.

Sent from my Nexus 6P using Tapatalk
__________________

- 1965 911
- 1969 911S
- 1980 911SC Targa
- 1979 930
Old 04-30-2017, 05:19 PM
  Pelican Parts Technical Article Directory    Reply With Quote #835 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
Quote:
Originally Posted by kenikh View Post
Yes. I forgot you don't have one. I installed one on my car.

Sent from my Nexus 6P using Tapatalk
I'm not sure thats relevant to the problem anyway. I'd like to hear what Steve has to say as he modified the firmware.

Maybe there's something wonky in ASE algorithm or something.
Old 04-30-2017, 05:27 PM
  Pelican Parts Technical Article Directory    Reply With Quote #836 (permalink)
3 restos WIP = psycho
 
kenikh's Avatar
 
Join Date: Feb 2005
Location: North of Exit 17
Posts: 7,714
That's what I was thinking, it is just another way to bypass the speed density algo.

Sent from my Nexus 6P using Tapatalk
__________________

- 1965 911
- 1969 911S
- 1980 911SC Targa
- 1979 930
Old 04-30-2017, 05:33 PM
  Pelican Parts Technical Article Directory    Reply With Quote #837 (permalink)
umop apisdn
 
Join Date: Jun 2014
Location: Brisbane, Australia
Posts: 445


There doesn't appear to be any correction going on internally, warmup and after start have no effect on VE2 and the FV in dual table mode, Accel/Decel and Ego do get folded into correcting the VE2 results but check "Fuel: Total Cor" in the logs and it's a nice flat 100% throughout


from the 5th start log

MAP, VE2 and therefore FV duty all drop a couple of points at 15.44 seconds in

AFR is flat at 15.4 for another .2 of a second and then drops to 12.5

from the 3rd start

MAP, VE2 and therefore FV duty all drop a couple of points at 13.20 seconds in

AFR is flatish at 15.8 for another .2 of a second and then drops to 12.5


The Wideband is still warming up so the lean indication may be a red herring but
it could also be a vacuum leak in the time based mechanical stuff somewhere

The MAP sensor has a very distinctive increase in vacuum at the transition point and the FV only changes by 2% so I'd be looking at the emissions and valve plumbing - what is active or open for 10-20 seconds after starting the motor

If the AFR reading can be trusted 15->12 seems too big a swing and in the wrong direction for a 2% decrease in duty, there has to be additional air sneaking in somewhere.

to eliminate the FV as a factor, check your VE2 table and try and flatten it out between 50 and 80 kpa and below 800 RPM so the VE changes very little and stays around 90/91
__________________
Steve
1981 928S 4.7 ROW with KE3-Jetronic and Franken8 (AEM Inifinty) follow at [http://www.frankencis.com/Activity-Feed/userid/2]
Yes! mechanical/hydraulic constant flow injection can be managed by a modern EMS

www.FrankenCIS.com

Last edited by Reanimotion; 04-30-2017 at 07:17 PM..
Old 04-30-2017, 07:13 PM
  Pelican Parts Technical Article Directory    Reply With Quote #838 (permalink)
Registered User
 
Join Date: May 2015
Posts: 616
Yeah forget about the AFR. The controller has a calibration sequence on startup and that is the sudden drop you are seeing. With AFR disconnected the problem still exists unchanged, which is why I performed the test with it hooked up - to eliminate it as a cause. Its neither responding to, or causing, any issues. A complete red herring.

"flattening VE2 around idle should go a long way toward eliminating some factors"

Flattening? or fattening? Can you explain a little further?

"cause and effect relating to MAP, RPM and VE and the resulting frequency valve duty is definitely in question, with MAP being the point of interest I think"

I'm convinced the change in MAP is an effect not a cause. What would cause a sudden change in MAP? The only thing I can think of is a change in RPM, which would obviously cause MAP to change, and what I think we're seeing here. The only other thing I can thing of that would cause MAP to change, in other words making MAP the cause not the effect, would be the idle valve. But that works perfectly as demonstrated by the non-FCIS starts.

So, what causes a sudden change in RPM other than the sudden addition or reduction of pressure in the intake (opening or closing a throttle or valve)? Well there's only two others options, spark and fuel. I'm still using the factory dizzy so the only possible way the timing can change is a change in vacuum or pressure on one or both of the ports on the vacuum capsule on the dizzy. Well the thermovalve has been eliminated and the advance side runs straight to the throttle body, and the Pierburg switch on the retard side can be seen operating on the cold starts, not the warm starts, and at a grossly different time than this stumble. I think that eliminates ignition as a cause.

"It the change an MAP and the resulting VE2 values changing the RPM, or is the cause resolving itself and subsequently changing MAP and RPM and therefore we get a new VE2 value."

I think its the latter. Something is changing to cause RPM and therefore MAP to change. The only thing I can see that is changing is the FV duty cycle, adding a bit more fuel. So the question is why is that changing?

"Is MAP fed from the manifold or is it tapped into an existing line somewhere which may be either ported or affected by the oem valves and switches? "

It is teed into the vacuum line from port IV on the throttle body to the WUR. AFAICT this port is below the throttle valve.

Last edited by flightlead404; 05-01-2017 at 06:40 AM..
Old 05-01-2017, 06:13 AM
  Pelican Parts Technical Article Directory    Reply With Quote #839 (permalink)
umop apisdn
 
Join Date: Jun 2014
Location: Brisbane, Australia
Posts: 445
Flattening,
the left side of your VE2 table around idle varies between 90 and 95 or so,
as MAP changes or RPM changes a little there is a resulting change in VE which is changing the Frequency valve duty a couple of points right at the 16 second mark where idle comes good

I too think the map change is an effect of something else in the system, but the resulting VE change when map changes could in fact be the cause of the bad idle

if we can get VE2 to not change when MAP changes that little bit at idle, that should behave exactly the same as the factory FV control and hopefully stop the RPM change when MAP changes
__________________
Steve
1981 928S 4.7 ROW with KE3-Jetronic and Franken8 (AEM Inifinty) follow at [http://www.frankencis.com/Activity-Feed/userid/2]
Yes! mechanical/hydraulic constant flow injection can be managed by a modern EMS

www.FrankenCIS.com
Old 05-01-2017, 07:24 AM
  Pelican Parts Technical Article Directory    Reply With Quote #840 (permalink)
Reply

Thread Tools
Rate This Thread
Rate This Thread:

 


All times are GMT -8. The time now is 11:52 PM.


 
Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2017 Pelican Parts LLC - Posts may be archived for display on the Pelican Parts Website -    DMCA Registered Agent Contact Page
 

DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.