|
Good description, Jim. I used Weber DGV32/36 downdraft carbs on my old MG's....I can relate. Just need to take a very close look at the 930 throttle body relative to where the ports are to fully understand.
|
Break through!
Mark, Welcome. Very nice car. I find your post very exciting! Your timing numbers are much more ideal than the US numbers we have been working with. We see about 10 deg with vac or boost retard and 18 deg mechanical. You see more like -14 with vac or boost and 12 mechanical. This gives you a greater jump in advance off idle, allows you to run more advance at cruse for greater efficiency, and still pull back a significant amount on boost . A better set up for sure. To me this info is HUGE! Several times I have stated I think a way to improve our systems is to increase the range of Retard available to us. I believe this can be done by modifying the stop behind the pot that limits the total retard we can get. Mechanical Advance can then adjusted to fit the ideal on boost goal. If you car is doing so, it is proof it is achievable. Here is how the combinations might compare-- Using stock US values adjusting idle to TDC. TDC at idle, -10 on first accel -18 on boost -28 cruse Your Euro car w idle set at TDC: TDC at idle, -14 with first acceleration, -12 on boost -26 on cruse. Using US mechanical w euro advance/retard: TDC at idle, -15 on first accel -18 on boost -33 on cruse Of course you can move all these values up or down by where the timing is set at idle. If we do this with the Euro to get the on boot to -18 it would look like this: -6 at idle -20 with first accel -18 on boost -32 in cruse. This might give a lot better off idle pull, safe on boost, and better fuel economy on cruse. Ya! I love it! The best!!! ;) |
Quote:
Crazy people like me put in 10:1 mahle pistons, 300 degree duration BMW motorsport or Shrick cams, Stahl tuned headers, and dual sidedraft Weber 45 DCOE carbs with 38mm venturi's while hogging out the intake ports to around 44mm after looking at and studying alot of pics of Jim Rowe's BMW M10/M30 engine intake porting theory in his Metric Mechanic book. |
Ignition Retard, CIS, Lean surge, resident boost?
Another light bulb might be going off. Carbs and EFI have a method to deal with the lean surge that comes with acceleration by increasing fuel flow. 3.3 930's do not. (stock 3.0 930's do.) If we run significantly retarded timing as Porsche specs at idle, the adjustment at the air bypass screw must be opened to keep the idle on target. This increase in air and fuel then lies in wait for timing to create acceleration by just quickly advancing the timing. With retarded timing exhaust temps go up as the fuel is still burning when the exhaust valve begins to open. Thus, less energy it imparted on the piston and more on the turbine. This should increase our resident boost. (I used to have about .2 bar off idle.) We have found that more advance increase pre-boost response. Could it be we want to keep our timing retard at idle and concentrate jumping advance more off idle. If so, for those going to a programmable MSD, you might try keeping idle timing close to stock spec and then get aggressive with timing advance with acceleration. |
In real life use just removing the red vacuum retard line from the distributor advances timing around 10 degrees at idle and makes these cars alot peppier at city driving speeds without boost so I don't promote the stock retarded ignition timing at idle.
There's no use for it other than the higher exhaust temps it causes at idle to help heat up the catalytic conveter and lower exhaust emissions. The 1987 930 owners manual even warns against letting these cars idle a long time while adjusting the idle CO because it overheats the intake ports from heat transfer from the exhaust ports and there is less cooling airflow from the fan. FWIW, apparently letting the cylinder head ports overheat will make it difficult to properly set the idle CO accurately. I've never heard anyone besides you talk about off this "off idle lean surge". I don't have an off idle lean surge with the modified fuel head in my car and I don't think my buddies '87 930 with a stock USA fuel head does either. We both have the Innovate LC-1 controller and analog AFR gauge to see whats happening. The 930 CIS takes a second to react to throttle changes with the low airflow going through it just off idle but from my experiences it's not a "lean surge". |
J,
If just removing the Vac-Retard connection (not removing and resetting timing) to the pot makes the car snapper then the power of more ideal advance overrides which is something I was wondering about. I do not like the idea of making any more heat than necessary. I do not know if 'lean surge" is an accurate term to describe my observation. It is not so bad the car coughs. On my 3.3 C2T with euro fuel distribution head I used a stand alone Motec wide band when setting up my controller driven Andial-Fueler frequency-valve to dial in my AFR's at all points off WOT. (WOT was done on a load dyno.) With aggressive throttle application I noticed my AFR did lean at first. If I was set in the mid 14's at idle -- I would see a jump into the 15's briefly with aggressive acceleration. I am surprised if this is not evedent to others running Wide Bands. If that is the case, I guess I could have had an issue others do not ( such as binding in the metering plate, thought I had it apart to blue print the metering pin travel and noted nothing). Another possibility might be the Motec unit was more sensitive to change. I know the O2 sensor was Mote supplied and new. Maybe there could be some level of soothing difference between the two instruments to make them more readable. I believe most of the sluggishness of CIS comes from the metering plate restriction and mass. Many CIS cars add enrichment (reduced control pressure) with loss of vac including VW, Audi, and Porsche. I was told the 3.0T did, 3.3's do not. My biggest single improvement to throttle response off idle was aggressively lowering control pressure with acceleration for a brief period. I am surprised no one else sees some leanness at first with aggressive acceleration. |
Wideband AFR gauge kits have gotten better over recent years so if anything the new ones are more accurate than the Motec you used when you had a turbo. I think you've said that was around 6 years ago?
My friends car and my car are both USA 1987 models and both have 964 cams. His car has a stock USA fuel head and both cars have a K27 7006 turbo and halfbay Garretson intercooler. Both of our cars idle alot smoother and faster with the AFR hovering around 13.8:1instead of 14.7:1 when fully warmed up and thats where we have them set. Maybe thats why we don't go lean into the 15's when first giving it the gas. My steady cruise AFR is around 14.7:1 at around 2000-2500 rpm and then it hovers around 14.5:1 at 3000rpm in 4th gear or 80mph on flat Florida roads. |
I don't see the leaness, Keith. If it's there at all, I'm talking a scant 1/10's or 2/10's AFR for a half-second while the metering plate moves, but not noticeable from a seat of the pants performance standpoint.
If we really wanted to test it, and since I still have my old Andial fuel enrichment gizzmo plumbed in, I could trigger it at will from the cockpit...say within the first second of touching the throttle, and force a rich condition. Or as I threatened to do before, put a window switch in to activate the Andial at 1200 rpms and turn it off at 2500 or some arbitrary point like that. The problem with all that is too many gizzmos hanging around the engine bay, doing what a nice programmable controller could do (such as you once did, and as Thierry has done, with the AIC). |
JF,
I have suspected setting the CO at 3% was to improve off idle TQ. Do the LC's have a sensitivity setting? Anyway, when I got done with my set up I idled at about 14.5 and with accel would jump to about 12.5. Again, I lowered CP to under 1.9 bar for a brief time with acceleration and did not know about 13.2 was a more powerfull setting. Same thing on cruse. About 14.5 and instantly to 12.5 then to 12.2 on boost per the advice of my tuner at the time (he was an import turbo guy). I told him I wanted to be in the 11's on boost but he talked me out of it. I think I ended up at just under 13/1 at read line at .92 bar (K27-7200) where I was running out of fuel and could get no more air out of the turbo. |
I think stock CIS provides acceleration enrichment through overswing of the sensor plate as you initially jump on the gas. This momentarily takes it past the full throttle open position. This is set out in the Bosch Technical Instruction for K-Jet (probably shouldn't post the extract).
On my logging I can't find an instance where AFRs go lean as the throttle is aggressively openned in gear. AFR's seem to drop in direct relation to RPM climbing at WOT. Perhaps it's not sufficiently granular but you can't see it on the trace. David |
There is no sensitivity setting on the innovate lc-1. It's a weather sealed unit.
To fully calibrate the lc-1 the first time or if it has become corrupted you have to clear it's memory by turning it on for 20 seconds without the o2 sensor connected and then turn it off. Then you connect the o2 sensor in free air with no exhaust fumes around it and turn it on for 2 minutes to calibrate the lc-1 to the o2 sensor and it's heater element. Then turn it off for at least 30 seconds before starting the car with it installed in the exhaust pipe. They say you should do a free air sensor calibration once a month on a turbo car, and before every track session in a track car. You don't have to do the memory clearing step every time if it is working correctly only the free air calibration and a little red button and red LED come with it for free air calibration. You don't have to remove it from the exhaust to do that either, just wait overnight and do it before starting the car once a month or whatever. You can update the firmware on it with a laptop if needed but thats more involved. |
Got it, the LC's can be calibrated but there is not adjustment for, shall we call it, refresh rate or soothing.
Maybe this is not even a function of O2 monitors as they do not react to changes in AFR's fast enough to need such a feature. It seems the way CIS is designed it is much more sensitive to air flow changes off idle and cruse. This is part of how it deals with potential for lean-onset, shall we call it. Some of the WUR are vacuum sensing so when vac is reduced with acceleration the reduce control pressure some to richen the AFR for acceleration for what it is worth. Seems getting our AFR's to about 13.2/1 pre boost (max TQ setting) and the timing advance right is going to be our best goals. |
Found some old info from from a very good source on Rennlist back in 2001 when I was setting up my programmable CIS Fueler and was looking into what my AFR's should look like that might be a couple of interesting timing data points:
Quote:
|
;)
|
Quote:
He said: "I run about 14.5:1 and about 37-38 degrees of timing". and he said "...run about 26 degrees of timing under full boost - 200kpa (1bar) except where my torque is peaked and there I run 24 degrees of timing. I aim for a 12.2-12.5afr". I know that information is specific to an EFI car. Personally with CIS, I run about 38 degrees advanced at 4000 rpm's no-boost/14.2:1 AFR, and about 16 degrees/11.8-12.0 AFR on full boost. Safe with neighborhood generic 93 octane gas. Interesting to see where additional timing can be had without ill effects, though I don't think I have the balls to go with that much on boost with the relatively crude CIS. |
Some great data points from a real expet I thought would be good to referance here:
Quote:
|
Here is where it cam from: http://forums.pelicanparts.com/911-930-turbo-super-charging-forum/537769-replacing-distributor-2.html
|
I appreciate Richards input. If I had the money he charges for his parts I could go to EFI.
Cole |
Let the street be your dyno and let the SafeGuard adapt timing in each cylinder to the conditions on that day.
|
john at J&S - is there some tuning needed to get the safegaurd working right? I thought the air cooled motors were so noisy that knock detection was difficult
|
Sorry but not again.
|
911st:
Yes, it's me again. Why not consider a system that gives you power, efficiency, and safety all at once? RC says the best way to do this is on the dyno. Engines blow up on the dyno all the time. How many dyno operators even listen to a knock sensor? If you tune by listening for knock, you are tuning to keep the worst cylinder quiet. I just ran across a couple old threads with some supporting comments from "350HP930". I don't know this guy from Adam but he loved the system. He hasn't posted in quite a while so I don't know how to get a hold of him: http://forums.pelicanparts.com/porsche-911-technical-forum/117549-930-how-can-i-richen-up-fuel.html http://forums.pelicanparts.com/porsche-911-technical-forum/120086-anti-detonation-system.html The unit he had was our first generation, produced from 1991 to late '98. Unlike his early unit, the current version incorporates boost retard with both start and rate controls. |
Quote:
The application I've used it on mainly is known to be the most knock sensitive internal combustion engine and it's saved me dozen's of times. Tuning on the dyno gives you a base line to test and setup. Actual real world conditions is where the car is driven/race at. Knock detection is needed when finding the limits during those conditions. |
Sorry for implying dynos are for blowing things up. That was a bit of a non sequitor. :)
I was trying to say that even a tune developed on a dyno can be improved upon with an individual cylinder knock controller. MoTeC, Autronic, and Link have recently added individual cylinder knock control. Why would they go to that expense if the tuner already has full control over fuel and timing maps? Imagine how much more useful the unit is on an engine that relies on thirty year old mechanical systems to control fuel and spark. crispeed: Thanks very much for your support. |
While not a 911, my 951 is equipped with one of John's Safeguard boxes and I sell his box bundled with my Vi-PEC EFI kit. The closed loop, individual cylinder knock control can't be beat. I have developed my ignition map logging the J&S activity to where when detonation occurs, my logs indicate the amount of timing retard the box applied to eliminate the knock. With all variables factored into the knock event - MAP, range of RPM, AFR, I have refined my timing map to within a few tenths of a degree of the "edge" of detonation. I leave no power on the table and I safety extract the most from my engine. I have no need for building a "safety buffer" into my timing map because the J&S makes individual cylinder corrections in real time without me even feeling it. Beautiful thing is that it still lets me know about it thru my logs!
Here is a screen shot of my J&S box in action. As an additional safety buffer, I have instructed my ECU to pull boost if the J&S is pulling more than 2 degrees of correction. The box will allow up to 10 degrees of correction without you even noticing it is doing anything. Boost is the black line and the purple line is knock actvity. http://forums.pelicanparts.com/uploa...1272576649.jpg For any of the non believers that are under the impression that the J&S or any other knock controller do not have the ability to distinguish between engine noise and knock, the great thing about the J&S is that you can use earphones to verify any signs of knock. The human ear is far superior to any knock specific electronic device, so all I had to do to accurately set the sensitivity adjustment on the J&S was to listen for the sound of slight knock (readily audible) and coordinate that to the triggering of the activity LED(s). Once the timing between the audible knock sound matches the activity LED(s), the box is set to its job! Eventually, we plan on offering Vi-PEC EFI kits for the 911 line without a doubt using the J&S as a key component. Thank you John for an AMAZING product. |
John,
I appreciate John's ingenuity and perseverance and I'm quite sure you have a fine product but I wish you would just start a thread of your own. It's getting to the point that if we had a forest fire thread here and the word "spark" was in it I would be reading about John's knock sensor all over again. Cole |
Ouch, the door hit me in the butt on my way out.
|
Quote:
Quote:
|
John,
Tact and sutulty have never been my strong suite. I promise if ayone asks about knock sensing I'll personally give them your name. Thanks !!! Cole |
I apologize if I have offended anyone here - I mistakenly only read the first page of this thread and thought my input may be useful. OP, let me know if I should just delete it.
|
Not a problem.
If you dig back in this thread I think you will find this disscussion, 944T v 911, motor noise interferance, the effort required to adapt a 911 to knock sensing -- was covered. |
Vic,
By all means if you have something to offer in reference to a post, that's what we are here for. I apologize for sweeping you up in this, not your fault and including you was wrong and totally my mistake. And believe it or not I don't dislike John, as a matter of fact I'm very impressed with him. I just want to go look at his site, which I have done, instead of getting constant commercial endorsements from him. I think John has a wealth of information to contribute to many threads that don't pertain to knock sensors but I don't see that. I see him siezing every opportunitie to plug his product instead of contributing to a thread and that was not, and is not what these forums are about. Wayne could go the cable networks route and flash Paid Programing in the middle of a thread and sell advertising. Cole |
I am wondering if we can adapt a late SC dist pot to a turbo?
It seems it has both the ability to advance with vacunm and as it has vac retard so it should let it retard with boost using only one conection. The stock dual pot dist only can retard. I am told Vac retard is about 7-10 deg (this would function as boost retard) and vac-advance about 5-8 deg (do not want this active at idle). That makes for about a 15 deg swing from light load at cruse to full boost WOT. I am thinking this might let timing would look about like the following: -12 deg static timing (no vac retard and cooler more efficent idle) 15 deg of mechanical advance (same as stock) -27 being the total deg of static plus mechanical. (WOT no boost at speed) On boost retard would bring it back to about -19. At cruse, vac advance would take timing to about -34. Then fine tune from there. The major gain is access to better efficency and a cooler running car at cruse. I am not sure we will be able to acces Vac Retard at idle though it might be possable. |
Probable Correction!!!
Euro Dist has Vacuum Advance, not Retard. During this thread info was presented that the single connection euro dist only has vac retard. It seems that is not the case. It seems to have Vacuum Advance with makes more sense. Looking at a picture of the euro dist bellow, it looks like the dist turns counter clockwise and the pot pulls, advancing timing at cruse. There is not effect at idle as it is attached to a ported connection that does noe see vac at idle. If this is true as I suspect then idle speed should not be effected if the connection is removed at tile. If it is a Vac Retard, disconnecting the vac line would increase idle speed. I like that if there is a rupture on a single euro that it should fail at a save timing value for on boost. The dual pot with a failure might not retard on boost which could cost a motor. On the other hand, I suspect a dist with boost retard will have a little better retard curve over a vac advance dist which will probably loose all its advance with loss of vac before boost is achieved. Just a thought. http://forums.pelicanparts.com/uploa...1254023296.jpg |
Quote:
David |
David,
Sorry, all have one pressure box or pot. The early 3.3 euro turbos, as pictured above, seemed to have a single vac line conection for vac advance. The later 930's, like you seem to have, seem to have the US style dist with two vac line conections. This style only has retard. Vac retard at idle and then retard on boost. |
sub
|
Had a little time on my hands today and wished to refresh my memory on timing, such as it is. Old rehash; but here's my attempt to put 513 posts into words. Do I have it all correct....I doubt it, but you decide.
Good Lord, 99% of this thread is supposition and creative thinking (better referred to as "learning"), whereas the remaining 1% may actually hit on some truths. If I were to attempt to distill it all down to what's happening with timing control for a dual-pot distributor, I would start by reading posts 74 and 138: __________________________________________________ __________________ ...tested both sides of the vacuum pot under pressure and vacuum. I found that the advance side of the dizzy responds quite a bit to vacuum and retards a little under pressure. The retard side only responds to vacuum. So at idle the timing is retarded by vacuum, when you push the pedal that vacuum is gone advancing the timing. when you accelerate the advance side sees vacuum and advances the timing with both vacuum advance and mechanical advance. when under boost the vacuum turns to pressure in the line, eliminating the vacuum advance and starts to retard the dizzy with pressure. Am I understanding this correctly? 911st mentioned a solenoid plumbed into the advance side, this is confusing to me. Which solenoid are we talking about. is it the one that looks identical to the one on the retard side, that is closed for about a minute to give you a high idle at startup? I thought the other one was used only for emmisions air pump etc. or are we talking about the thermal valve up by the crankcase breather hose. __________________________________________________ __________________ ...put around 15" of vacuum into the left side or outer advance pot that the red line goes to the engine speeded up. When I did the same to the inner or right side retard pot the blue hose goes to, the engine slowed down.. That means the left one is advancing the timing when vacuum hits it and the right side one is retarding the timing when vacuum hits it and my guess is they cancel each other out when boost pressure replaces vacuum in the 2 lines leaving only centrifical advance which may be what boost retard amounts to..... __________________________________________________ __________________ The truths: Vacuum at the retard side of the can (closest to the distributor) AT IDLE will cause retarded ignition. Open the throttle a little and that vacuum goes away - because of where the port is located in the TB - causing the ignition to immediately advance back to static timing....or where it would have been had it not been retarded due to the vacuum. Now push open the throttle plate a bit more and two things happen: You gain some amount of vacuum advance through the advance side of the can - because of where it's port is located in the TB....and also gain mechanical advance as the rpms increase. Now put your foot further into the throttle and you transition from vacuum to boost; i.e., your vacuum (and thus your advance) goes away and all you've got left is the mechanical advance portion. Net-net, your timing just got retarded when the vacuum advance went away. Under boost, both sides of the can see "pretty much" equal pressure and negate each other but you still end up with the same result....the absence of vacuum advance and nothing but mechanical advance left. That alone drops your timing probably 10 degrees to protect during boost. As for the two solenoids repeatedly in question (one electrical and one mechanical thermo time) the mechanical one hidden behind the fuel distributor blocks the vacuum advance when cold, and the electrical one blocks the vacuum retard when cold...but nobody seems to know how it gets it's signal to open or close. Quite frankly, neither of those two gizmos are needed and just serve to heat up the exhaust gas for emissions purposes. |
Quote:
It's in line with the boost signal to the distributor that causes boost retard under boost... I understand the same line does vacuum advance once warm... I see a lot of folks remove this switch... Any consequence to doing that? Higher advance under idle when cold? Does it making starting difficult? When cold, air doesn't blow through this valve... Either boost or vacuum. Does this mean you don't have boost retard until a motor is fully warmed up? Did most of you take this out of the loop? If this switch fails, doesn't that mean your motor grenades as you then lose boost retard? |
THIS IS FOR 1986 TO 1989 US 930'S ONLY!
The Thermoswitch cuts off vacuum to the outside connection on the distributor when the engine is cold, raising the engine idle speed, which is nice for charging, warm-up, and general drivability. The presence of vacuum retards the ignition timing. The absence of vacuum allows it to be advanced. The other distributor vacuum port, on the inside, is responsible for boost retard. |
| All times are GMT -8. The time now is 05:56 AM. |
Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2025, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2025 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website