Supercharger install

Zitat von fuzzifikation;76162

One thing about lag - if I may

Turbo Lag is the time a turbocharged engine takes to produce significant power at a fixed engine speed. So there are different lag times at different engine speeds. Correct, an example would be, for anyone having not driven a turbocharged petrol engine, at 3000 rpm at a light throttle setting, ie constant cruise speed, the turbo may not produce any boost but stamp the throttle and after the lag period, the car may well produce 15 lbs of boost, virtually doubling the horsepower, still at 3000 rpm. (A very slight increase in engine rpm will obviously occur whilst you are waiting for boost to arrive but minimal for the purposes of this example)

but.. there is one other thing:

Turbo Threshold is the minimal engine speed at which a turbocharged engine starts to produce significant boost (independent of how long it takes it to do that .. the time would be its lag at that speed). Again correct, similar to a centrifugal supercharger except from that point, the turbo boost will be throttle position dependant and the supercharger boost will be rpm dependant, although not linear from that point as the supercharger compressor enters its efficiency map.

So:... A supercharger (and especially this turbo compressor which, obviously has been put in this engine - forgot about that) has threshold similar to a turbocharger as it will not produce significant boost at very low rpms... We agree!

However it will still produce more boost at low rpms than a turbo which has been chosen to produce the same driving characteristics. That is because it does not have a turbine which simply worsens the overall efficiency and will not run as well down there..We disagree! The turbo turbine is effectively an exhaust system blockage and anyone who has experienced a turbo bearing failure will tell you that a turbo engine will make more power with the turbo completely removed than one with a failed turbo. I carried out a bodge repair many moons ago by removing the shaft and impellors from a failed turbo and putting a nut and bolt through instead to seal the oil way and the holes left by removing the impellors and it drove very well. At low rpm, it will not have reached its efficiency maps so will not make boost, but the fact that it is spinning in the correct direction will reduce parasitic power draw. A turbo also spins in the correct direction but this is due to the exhaust flow over the turbine and the induction flow across the compressor. As my example, put a small kiddies wind fan in your intake and the air flow over it will cause it to spin. This is absorbing flow from the intake to power it. If you manually spin the fan, then no power is absorbed by the fan as it is running at that speed anyway. No power draw. Drive fast enough and your electric rad fan does nothing as the airflow across it spins it without any help from the fan motor.

With the other stuff.. I'm afraid that Gerry does have a point. A turbo compressor will draw little power from the engine if it's not actually compressing air a lot. Thank you.

But: If it starts to compress air even a little, its power consumption is quite bad. This is, because it will operate at very low efficiency until it reaches a higher speed and high compression ratios. As soon as the compressor starts to compress the air, then it will take power to drive it but with a 98% efficiency on the drive system, there is virtually no parasitic loss off boost, where you will be most of the time.

And then there's one more thing... if a supercharged makes an engine more efficient (which it does to the compustion efficiency) - it does this only by adding boost to the intake manifold. Which needs power.

Finally: A turbocharger is wildly more efficient than a supercharger considering overal fuel economy! Nope, sorry, disagree there.

Why? Because with a turbo, the control unit (nowadays they're pressure controlled by electronics) will simply open the wastegate if power is not needed. Then, the turbo will not produce high boost, whereas the supercharger always does.
Worse so: If the driver doesn't need the boost deliverd by the supercharger (he steps off the gas pedal) then the throttle plate will need to eliminate (= waste) all the precious boost that was created by the supercharger (at high cost!). If you want to compare the latest technology then fine, the manufacturers have developed turbo's at great expense beyond supercharging, but we are discussing the aftermarket here. The same technology could be used on a supercharged engine, instead of dumping the boost, recycling it back into the intake.

Note that when I'm comparing turbo vs. supercharged, I don't mean those turbocharger approaches with fixed boost - these are only ok if you want to race your 8er all the time - it's just bullsh... for everyday driving. But that is exactly what we are discussing here, a performance turbo set-up, not a factory install.

A supercharged approach is always better than a turbocharged approach that is done wrong (fixed boost). HOOOO_RAY!!!

Gerry.. your turn

Alles anzeigen

Back to you!

This is much more like it & I can actually understand what is being said

So... supercharger options on an 850CSi?

Zitat von mattsimis;76190

So... supercharger options on an 850CSi?

You seen this: http://www.8er.org/forum/showt…ight=supercharger+project

This is fun!

Zitat von Argonaut;76196

You seen this: http://www.8er.org/forum/showthread.php?t=6228&highlight=supercharger+project</span></a><span style="color: #800080"></span>

Yes, I know that Gerry knows what he is talking about... however, I am not totally unknowing about this subject, either (having spent many years on turbocharging at work.. actually I'm working on yet another projekt now).

Now... IF we say that we turbocharge an engine with a 1980's fixed boost setup.. well... that's crap. Supercharging is WAAAAY better than this! Unfortunately, aftermarket turbos often have exactly this, simply because a fixed boost setup allows that the turbo needn't to have a computer that is controlling it and would have to talk to the engine DMEs, also. So, in essence, a modern turbocharging approach is much, much more work than what Egil goes through in this thread.

But it's also better, fuel economy wise.. But this is not the reason each and every car manufacturer, nowadays, is almost exclusively producing turbocharged engines. The reason for this turbo-inflation is, by the way, the ability to immensly increase exhaust back pressure which can be then used to realize high EGR rates, which are necessary for the EURO6 emmission norm..
And not only that! A controlled turbocharged setup lets you freely choose the torque behaviour of the engine at any speed! This is the reason for all of those perfectly flat torque curves of the new engines (e.g. the 535i) - > a supercharger can't do that.

However: with the premission that we are comparing a 1980's turbo setup with a supercharger.. I'd pick the supercharger simply because of its much better driveability, as a blind turbo (fixed setup) gives you nothing better than the supercharger at the cost of lag.

As far as fuel economy goes: we disagree! A supercharger needs (and wastes) engine power - and not little of it. Even with 100% Efficiency. As soon as you want boost from it you are immediately in the range of 20..30..50..70 HP for the supercharger alone. (I will do the math whan I'm back at work as I am too lazy to get the formulas out of the car)

And this is, what I have trouble believing: 98% Efficiency ? With a turbo compressor? Really? From what I've seen, turbo compressors top out at around 80-85 (very few closer to 90)% and drop (in normal driving) to as low as 40%. The problem here is, that efficiency is not a constant number with these machines - it varies with the speed of the charger, the temperature, the intake and output pressures..

If a company would actually be able to produce 98%, they would immediatly destroy all other turbo manufacturers. What company is that and is there a spec sheet available?

Maybe something like this: (and here, again, we top out at 78% efficiency with a pretty good turbo compressor - which by the way will NEVER manifest in driving.. only during testing conditions on a gas test stand will you get 78% with this on below - in a car.. maybe 73%)



Oh .. and : with a turbo there is very little friction since it sits on an oil bearing + (sometinges) a roller bearing. That stuff is very, very low friction. What's the bearing on the supercharger? Also oil? And the friction on oll the belts, pulleys, etc? Well.. that stuff adds up!

I guess we will agree to disagree. I am absolutely convinced that - unless you use boost to drive at lower RPMs - you will waste fuel noticably!
(but then again.. most people to exactly that: drive with lower rpms because they love that low-end torque feeling)

Servus Dieter,

don't misunderstand me, I wasn't trying to pick holes in your knowledge, I merely provided the link to Gerry's project for the benefit of Mattsimis in case he hadn't already seen the write up. You and Gerry obviously both know what you're talking about, I'm just an interested bystander in what is a very interesting debate. With my limited knowledge there's not much I can contribute to the proceedings but I find it very interesting. Personally, from what little I know of the subject, supercharging makes far more sense that turbocharging although, I think you would have to take into account the advantage that a turbo has in making some use of what would otherwise be simply wasted energy in the form of hot, fast flowing exhaust gases whereas a supercharger is reliable on pure rotational torque power from the engine itself to develop boost. Surely this would factor into the efficiency equation in some form?

Cheers,
Jason

Hi Jason! No, no.. I just like a debate! (and I hope the others here, do too.. if that shouldn't be the case, I will happily shut up! - I'm not trying to annoy anyone - just like a debate! :-))

On your question about exhaust gas energy.. yes, of course that factors in. Withouth that, we'd all be using superchargers as they are so much simpler to handle (less space, no lag, so much less parts to manufacture and so many fewer hoses). The whole point of the turbocharger is to use the exhaust gas energy instead of taking that energy from the crank. I'm sure that nobody denies that this - in it's core - is a great idea! Although it's - again - not quite as simple since you increase exhaust back pressure with the turbine. That also lowers engine power output directly (but not quite as bad).

Going into the details of that is quite difficult as properties are not as easy to predict since everything depends on how different parts are chosen and how they interact with one another.

My whole point is this: A supercharger will severely increase your fuel consumption unless to start to drive your car with much lower rpms.

However, most people do start do drive it in lower rpms, when they experience the brute force in the low-end. Usually they love it! And then they start cruising around at 1300 rpm instead of 2000 rpm and bingo! They save fuel!

Oh.. and of course my other point is: If you cruise at 3000 rpms with a supercharger you will need a lot(!) more fuel than cruising at the same rpm with a (non 1980's) turbo.

So.. that's my whole point

(and, maybe, that no turbo compressor has 98% efficiency in any point.. sorry, but I am willing to bet a keg of beer that 98% is a blunt lie from whoever is manufacturing that thing! :-)).

Zitat von fuzzifikation;76210

Hi Jason! No, no.. I just like a debate! (and I hope the others here, do too.. if that shouldn't be the case, I will happily shut up! - I'm not trying to annoy anyone - just like a debate! :-))

My whole point is this: A supercharger will severely increase your fuel consumption unless to start to drive your car with much lower rpms.

However, most people do start do drive it in lower rpms, when they experience the brute force in the low-end. Usually they love it! And then they start cruising around at 1300 rpm instead of 2000 rpm and bingo! They save fuel!

Oh.. and of course my other point is: If you cruise at 3000 rpms with a supercharger you will need a lot(!) more fuel than cruising at the same rpm with a (non 1980's) turbo.

So.. that's my whole point

).

Alles anzeigen

But my previous statement contradicts everything you say

My car was Supercharged, 400 bhp 580 newtons torque, 3.81 Ls Diff, so it would not be possible to drive as you state in low rpms, on the contrary, it was so fluid at lower rev band i found i was screeming around every where,& yet my consumption was much better than pre conversion due to what I was told "the engine is working more efficiently on a leaner fuel/air mix";-)

Zitat von fuzzifikation;76210

(and, maybe, that no turbo compressor has 98% efficiency in any point.. sorry, but I am willing to bet a keg of beer that 98% is a blunt lie from whoever is manufacturing that thing! :-)).

I'll bet you a keg of beer that the drive system for Egil's supercharger is more than 98% efficient !

BTW, I believe you've mistaken supercharger drive system efficiency for compressor wheel efficiency. If you re-read Gerry's post, you will discover that he was referring to parasitic loss of the drive system (not compressor efficiency).

Insofar as compressor efficiency, the unit in Egil's car has efficiency range of 70-76%.

Hope this helps...

Zitat von stevep840;76075

The Old superchaged 840 of mine went from averaging 22 mpg to 26 mpg after the S/C Maybe I should have kept it.

Were these numbers taken from the car's on-board-computer?

Zitat von John in DC;76212

BTW, I believe you've mistaken supercharger drive system efficiency for compressor wheel efficiency. If you re-read Gerry's post, you will discover that he was referring to parasitic loss of the drive system (not compressor efficiency).

And you would believe correctly! Yes, I was thinking the thermodynamic efficiency was 98%.. which sounds a little high
I believe you that you have a very good drive system in there! 98% sounds perfectly reasonable. Actually, it sounds like you guys did a great job on it.. and - again - I believe that.

Zitat

My car was Supercharged, 400 bhp 580 newtons torque, 3.81 Ls Diff, so it would not be possible to drive as you state in low rpms, on the contrary, it was so fluid at lower rev band i found i was screeming around every where,& yet my consumption was much better than pre conversion due to what I was told "the engine is working more efficiently on a leaner fuel/air mix"

Well... you have significantly reduced your A/F mixture to lean. That's fine and yes, you will save some fuel with a lean A/F mix (also you will not have any functioning catalytic converter and less power than what you could have gotten, but I believe that to be of secondary concern).

BUT: Reduce the A/F mix on a non-supercharged same engine and you will end up with even more fuel savings..

or, stating it differently: You're comparing apples and oranges.
Get an engine, put on a supercharger and change the maps to a proper A/F mix and you will get much more power, a ton of fun and a smile on your face. But you just won't save any fuel... really.

Although I'm happy for you that you had such great fun with your car...

P.S. (getting back to the thread topic): Running Egil's car lean is not such a great idea. His car is based on the old M70, which does not have knock sensors. If you boost an engine you will increase the danger of knocking. Give it too little fuel and you increase that danger even more. If anything Egil should run its engine on the fat side (but I'm not a specialist on chip tuning, so I will end it right here and leave it to John and Firebird).

Zitat von John in DC;76213

Were these numbers taken from the car's on-board-computer?

Yes always during the 5 years of owing the car, ie before & after conversion hence the quoted figures.

Here's a question for John and Egil (and potentially Firebird, if he's reading).. (and so not to annoy the rest of you with going on and on on the discussion above):

I understand how you can arrange lean or fat injection maps. However, as I understand it, ontop of the injection mass in that map, there is the oxygen control (i.e. the control loop that tries to keep the oxygen value in the exaust (via the oxygen sensors) at a stoichometric mixture). And don't the DMEs even store correction values for the injectors based on that oxygen control loop?

If that were to be so.. how can any car with oxygen sensors and such a control loop (supercharged or not) run lean while cruising?

Let me specify my question a little more: I do understand that since the oxygen sensors are really slow to correct their measured value, their correction cannot figure into quick accelerations. There, only the mass flow sensors are used (as only they are fast enough to detect the air mass flow and the injected fuel mass is taken from the tables with a correction value that might be wrong).

However, cruise long enough (maybe 5 seconds or so) then should the oxygen control loop correct the fuel mass? And shouldn't then als the correction values be stored?

I'm really confused about this thing..

The only explanation I can come up with is that the injection timing has been left unaltered. That is: maybe (and just a big maybe) the injection is taking so damn long (for lots and lots of fuel) that it's continuing although the intake valve has closed already... but that sounds like a lame idea, even to me...

Zitat von Argonaut;76196

You seen this: http://www.8er.org/forum/showt…ight=supercharger+project

Yes, an excellent thread. What Im most curious about is not the selection of superchargers (Gerry made a great case for Centrifugal and the Wiki on it almost appears to be written about its application on our specific engines!), but how the stock and ancient engine management will modified to cope with the boost...

Zitat von fuzzifikation;76222

Here's a question for John and Egil (and potentially Firebird, if he's reading).. (and so not to annoy the rest of you with going on and on on the discussion above):

I understand how you can arrange lean or fat injection maps. However, as I understand it, ontop of the injection mass in that map, there is the oxygen control (i.e. the control loop that tries to keep the oxygen value in the exaust (via the oxygen sensors) at a stoichometric mixture). And don't the DMEs even store correction values for the injectors based on that oxygen control loop?

If that were to be so.. how can any car with oxygen sensors and such a control loop (supercharged or not) run lean while cruising?

Let me specify my question a little more: I do understand that since the oxygen sensors are really slow to correct their measured value, their correction cannot figure into quick accelerations. There, only the mass flow sensors are used (as only they are fast enough to detect the air mass flow and the injected fuel mass is taken from the tables with a correction value that might be wrong).

However, cruise long enough (maybe 5 seconds or so) then should the oxygen control loop correct the fuel mass? And shouldn't then als the correction values be stored?

I'm really confused about this thing..

The only explanation I can come up with is that the injection timing has been left unaltered. That is: maybe (and just a big maybe) the injection is taking so damn long (for lots and lots of fuel) that it's continuing although the intake valve has closed already... but that sounds like a lame idea, even to me...

Alles anzeigen

Hmmm, a good question Dieter, I also wonder why. My understanding would be that you wouldn't actually want to run at stoichiometric except when the engine isn't under much load since at that mix it will burn very hot. Nevertheless, I think you're right that with a closed loop system, the EML should correct the mix back to close to AFR (Lambda 1.0) so, if the map is controlling fuel only then I agree with you, the only explanation is ignition timing. (or simply that the oxygen sensors have been left out )

Gald to have asked a good question, for once.

Here's one more thing to - maybe - consider.
I have recently learned that the oxygen sensors are calibrated to work with within the range of specific pressure ratio. As you all know, the exhaust gas needs to flow into the oxy-sensore (from the side) and exits the sensor at its tip. If - for whatever reason - there is a pressure configuration such that this flow is not as it was measured with the factory setup - the oxy-sensor will/might deliver offset values.

So.. maybe (and again, I'm not familiar with the DME programming at all) it's not only ignition timing and injection rates to adjust.. but maybe also the oxygen value (lambda) with in the DME... ?

From how things appear to me, as of now, that would make sense... and it would mean that no matter what map - the engine might always be on the lean or rich side (I don't know which way the oxy sensor drifts) because of the oxy-control loop (which keeps the engine lean instead of rich/stoichiometric...)

Also, I don't know in what engine range the oxy-control is functioning (on) - hell.. I tried to obtain the original DME programming, but without success..

I seem to remember it being stated that at Wide Open Throttle (WOT) the controller goes open loop again. That it has a hard coded map for WOT and the O2 sensors are ignored.

So running down the road the system is closed loop after it warms up but reverts to open loop at full throttle such as on a dyno.

Don't remember exactly what device it triggers on --switch under the pedal or computer signals.

Zitat von fuzzifikation;76245

... and it would mean that no matter what map - the engine might always be on the lean or rich side (I don't know which way the oxy sensor drifts) because of the oxy-control loop (which keeps the engine lean instead of rich/stoichiometric...)

Here's a data log from when I was tuning my 850 Twin Turbo a few years back. Notice the square waves from the oxygen sensors.



Anyone care to guess as to why they are going up-down, up-down ?

Rgds...

Hi John,

I actually can explain that. That is, if your x-axis in the plot you showed is in the range of seconds (and not minutes or hours).
Your plot beautifully shows how an oxygen control loop works. And, before I start rambling on about why that is, let me give you a quick conclusion:

If your plot shows some seconds of engine running, then the oxygen control loop (or lambda control as it is called in Germany) is working correctly. (More so you can say that if you measure a constant line from the oxygen sensor.. the control loop is not working.)

Unfortunately, we cannot conclude wheter the oxygen sensors themselves give the correct values. But the pure functioning is ok.

Now, let me please ramble on how an oxygen control loop works.

First: There are two kinds of oxygen sensors. The newer one is a linear sensor meaning that you can easily conclude how much oxygen is actually in the exhaust (sometimes calles LSU - but that might be language-specific.. so I will not give any more names.. wikipedia will be your friend). But we don't have that one (for sure not in the M70.. probably not in any one of the 8ers)- so I will not get into that as the control loop also is slightly (but not much) different.

The older kind of oxygen sensor is quite dumb. And that's the one we see here! It can only tell you if you are running lean or rich. So, deviate a little bit away from stoichiometric ratio and the sensor will very quickly jump to extreme voltages.

The problem with that is, that the DMEs see only rich or lean - they virtually never "see" the nice stoichiometric voltage. Even if, they see that voltage, it's only for a brief time and then the sensor "jumps" again - simply due to small deviations. Plus, they don't see how rich or how lean the combustion actually is. That's aproblem.

So the car manufacturers went about it that way: They don't try to stay at the stoichiometric ratio all the time. Instead they WANT to toggle between lean and rich all the time. That's the simplest way to get approximately a stoichiometric combustion with that crappy sensor.

So, what you see in the plot above is exactly that behaviour. And, that does not mean that the combustion is changing wildly between lean and rich. No. It just means that the cumbustion changes between a little rich and a litle lean . It's the sensor that "acts out". Back at that time the linear sensor didn't exist yet - and if my memory serves me correctly, the linear sensor is just very fey years in the market.

(more correctly: it's changing between what the sensor believes is rich or lean)

However, as car manufacturers quickly found out using the linear sensor: This constant changing of lean and rich is actually a good thing. What it does it constantly "pumps" short waves of oxygen into the catalytic converter. And it turns out that this leads to a more robust conversion of the emissions. You see, a catalytic converter is like a moody woman. Give it just a tiny wee bit too much over a longer period of time and it stops working. Give it a tiny bit too little over a longer period of time and it stops working. But: Give it sometimes to much and sometimes too little - you'll get it's attention and things stay "fresh"

Again, the conclusion is: Things are as they should be.
That is if that plot shows some seconds and not minutes or hours.

Now on how to test an oxygen sensor:
If you change pressures in the exaust a lot, the sensor will continue to deliver voltages but its jumpy point might drift a little.... maybe to lean, maybe to rich (I'm not a chemist - other people in the company I work for do the chemistry of the sensor (as we also produce them.. actually invented them - but not me.. I have very little to do with that sensor.. I'm working on the diesel equivalent: a NOx sensor)))

You will need professional testing tools at the tip of the exaust to determine what's what (or a CO sensor right at the exaust manifold.. they have extra holes with threads for them). Then you will know if you are running lean or rich.

After that, you will have to do something about that (if you are running lean or rich.. if not - of course you're done).

In essence you might have to change the output voltage with an electric setup to change the jumpy point. That electric board would have to read the voltages and oh so slighty change them until you are happy with the combustion. Depending which way it drifts a simple resistor setup, combined with battary voltage will do the trick to shift it. Start with a potentiometer as resistance to be able to find out the correct values.


Did that help? Or was it something completely different that you've asked?