Centrifugal vs Positive Displacement Supercharger (Question)

[97M3POWER]

After reading a bit of Shawn Fogg article on Centrifugal vs. Positive Displacement Superchargers, I had a question! :95

Understanding that a Positive Displacement Supercharger produces a linear boost rate, say 8psi at 3000+ RPMs and also understanding that a Centrifugal Supercharger will produce boost at around the "square" of it's driven speed at any given RPM rate peaking out at REDLINE.

I quote Shawn here: "This means that at every point below redline you will NOT be at maximum boost. So say you had it set for 8psi at 6500 rpm. Then you can figure that at 3250 rpm (half SC driven speed) you now have the square root of the boost at 6500 rpm... or 2.8 psi."

Is it possible to affix a EBC or MBC (Electronic or Manual Boost Controller) as in a turbo vehicle say 18" or so away from your intake inlet, dialin 8psi and run a LARGER pulley which would produce say 8psi at 4000 RPMs?

Would this turn the Centrifugal Supercharger much too fast (Essentially BLOWING IT UP) at high RPM's? It there a middle ground?

[Kevlar]

I think it would

A) spin the supercharger so fast it would be out of it's thermal efficiency range causing it to just super heat the air rather than compressing it to go into the engine.

B) increase drag on the engine causing you to loose HP before making any

C) cause the supercharger to explode.

but then again, i'm just takign a stab in the dark here... I've been asking this same question for a while, but I think one of the major problems is...

they can't find a supercharger that can effectively spin the 20psi at redline (20psi at redline or 8psi at 4000rpm) that can fit in the tiny space we need it too... plus there is that whole drag thing too.

[97M3POWER]

I really would not like to see option C, :D

How about that happy middle ground that I was inquiring about?

Might we be able to spin one of these small footprint Vortec's up to 16psi creating around 4psi at 3500 RPMs and bleed off boost to 8psi at 7000+ RPMs?

If this were possible we would be near double the usable boost at low RPMs.

[Shawn Fogg]

"Is it possible to affix a EBC or MBC (Electronic or Manual Boost Controller) as in a turbo vehicle say 18" or so away from your intake inlet, dialin 8psi and run a LARGER pulley which would produce say 8psi at 4000 RPMs? "

You might spin the blower too fast which would of course put more stress on it. And as was mentioned the extra heat and drag wouldn't be very good. Also the action of something like a blow off valve would be very pulsey which I'd think would be very hard to tune around.

I've seen a few different attempts at basically mating a sort of CVT to a centrifugal (Paxton did this like 20-30 years ago) to try to get a better boost profile. I don't know how well that worked. Since you don't see it very often I assume not well.

Shawn

[psk145]

SHO Shop does that with their supercharged SHO's. THey put a restrictor on there that limits boost, but spool them up faster. They wanted to do that to mine, but I said maybe some other time...

[97M3POWER]

How did the SHO Shop implement this "restrictor"?

Was it like a BOV with it's internal spring set to a specific PSI level, say 8 pounds before opening and bleeding off air?

[psk145]

no, its like a fitting that goes into the intake tract that reduces the diamter of the piping in that area. Supposedly boost is limited, cause it cant get more air.

[Kevlar]

ya, so instead of having like a 3.5" intake pipe all the way in, you have a 3.5" intake pipe with a section with say 2" pipe. The 2" opening can only flow X amount of air ... reducing the air going into the engine to say 8psi.

But now... since the compressor is trying to push lets say 20psi of pressure through this 8psi hole... doesn't he air get super heated? doesn't the pressure build up affect the supercharger effectively slowing down the blades and adding increased drag to the engine?

[psk145]

Originally posted by Kevlar

But now... since the compressor is trying to push lets say 20psi of pressure through this 8psi hole... doesn't he air get super heated? doesn't the pressure build up affect the supercharger effectively slowing down the blades and adding increased drag to the engine?


yeah, that was the downside I was thinking of myself. I'd rather not worry about additional heat to gain some more low end boost.

I guess the w/a injection could fix all that though eh? :D

[97M3POWER]

Why restrict air coming in.. Let the compressor do it's job properly and bleed off extra boost once produced before entering the intake? *Right* :dunno

[stimpee]

I have spoken to several people in my efforts this past year that have actually implemented "boost control" on a supercharged car. In my case, they had done it with PD blowers. The PD blowers only come in limited sizes. For example, the compressor that will work on the BMW 2.5-3.2 engines is slightly oversized for the 2.5. The drawback of that is that the VE of the compressor is relatively low at low compressor rpm, which will slightly reduce the low rpm boost of the compressor. People in similar situations have slightly overdriven the compressor to make more boost earlier on in the rpm band.

The drawback is that you are compressing the air, and then bleeding it "off" or reducing the pressure. This, as mentioned, uses more power to drive the blower, and makes heat. HOwever in some cases, as long as the excess heat is not excessive, it may be a good trade-off.

In my case, I have considered implementing this in my setup once I get it working. My plan is to get it working simply first, and then possibly implementing boost control later on. Another cool thing about this, is that you could have cockpit adjustable boost with a SC. I could set it up for say 7psi for most of the time, but set the pulleys to get 10-11 psi peak. Then, with WI, high octane gas, or when at the track when you are really paying attention to the car, you could up the boost selectively.

I have a couple of other tricky ideas up my sleeve too, but I will save them for later. I need to get the basic setup working, and more importantly tuned, first...

Steve

[97M3POWER]

I guess if one would though water injection into the mix, the excess heat would kinda get canceled out. *Just how to make a "Bleeder" before the intake is the question.

[kitwetzler]

This would be really really easy to do.

All you'd need to do is use a typical bypass valve setup and use an adjustable restrictor valve in the bypass valve signal line. This would allow the boost side of the diaphram to see less boost and it would bleed off boost.

hm. this might make sense for us silly high mileage people... 11 psi pulley with a high and a low boost mode. :)

The problem is that you are still using a centrifical blower... at 3500 rpm you'd make 3.3 psi instead of 2.8 psi. Not a whole lot more. :)

That said, it'd certainly make your 5000 rpm+ very interesting.

-kit

[97M3POWER]

Yep.. 3500 would not see much improvement you said, but 4500-5000 would see close to a couple of pounds in excess!

I think the goal here is trying to make our SMALL foot print Slimline (Vortec's) come close to the effectiveness as it's "BIG Brother".

A) more boost at low rpm's it a better for the vehicle (not having to hit redline to see a full 8psi) no need to get there if you are at full boost by 6000.

and

B) would make for a stronger stoplight to stoplight street car.

[stimpee]

With the UNICHIP I can set up boost control using a PWM solenoid on the bypass valve boost line. That is how I would plan to implement if I were going to try it. In the case of the PD blower, it would give me full boost 500-1000rpm or so earlier in the case of the smaller engines. It would actually give me a full-boost range or "spoolup" that would make the ball bearing VATN turbo guys quite jealous!!!

Rubber could get expensive though!!!

[97M3POWER]

stimpee : You might want to look in the SMT5 system.. I spoke to an installer of both today and he seems to think the SMT5 is a bit better for our applications. SMT5 does store 2 seperate configs which is also nice. Let's say a 93 octane MAP and a 104. Great for track days.

[stimpee]

I have looked at the SMT5, and although it is cheaper, I have a few issues. First is that it has significantly less map resolution than the UNICHIP. The UC is 13 load points, and 17 rpm points for 13*17 (221) positions on the map. The SMT5 is a lot less (128), although I don't remember the load vs rpm dist. of it.

Also, I have seen the UC installed and working on OBDII cars, and it is very popular in Porsches which use the same or similar engine management system to our BMW's. So, until I see the SMT5 in more applications, I would probably have to reserve judgement. ALso, in my case I get direct support from The Racers Group and Dastek, so I am more likely to be able to work through any issues.

I am starying to think a bit more about whether I should maybe use the UNICHIP to drive some aux injectors rather than using it to modify the HFM signal to the ECU. That way it won't affect load calcs...

Hmmm, gotta talk to NickG about that one!!!

Steve

[97M3POWER]

That was one of the points that I made to the installer today. "Resolution" 410 to a smaller 256 points. He said 3 things. 1) The UniChip goes in 500rpm increments

... The SMT5 allows you to set you increments to whatever you would them to be by changing scope. Example. Let's say we want to only use our piggyback system to during boost and we no that boost hits at 3500rpm (example) no we can set 3500 as ground zero and 7500 as redline and use all of our point with in that scope. Also, lets say you know that from 6000 to 7000 is critical fueling area, you can set point lets say 250 apart from 3500 to 6000 and then 100 or 200 rpms apart for the those critical 1000 rpms. (basically having 5 point just between 6-7k if you wanted)

2)The UniChip has one setup (MAP)

...The SMT5 allows the setup of 2 and can be switched on the fly form one to the other.

3)The UniChip much be "tuned" by a person like himself that has a tuning device. (Not sold to the public) **I was led to believe**

...The SMT5 can be tuned by you, REALTIME in your car with a laptop.

MAJOR PROBLEM in my eye. so far. with both. NO REAL DATA LOGGING capabilities.. He says only the sensor that you are pluging into get logged, no water temp, no knock, no maf, etc. Meaning we would still need a Pocket Logger / OTTOSCAN or something of that nature to figure out where to add/sub fuel and timing from. Unlike my EFI, Inc (PMS System) where data logs over 250 ECU data points and also allows for realtime tuning while data logging realtime at the same time! GOD I wish DOUG would make that product for the Bimmers. You guys just don't know! Trust me! Imagine Haltec without the aggravation of standalone! :D

[stimpee]

I talked to Doug back in December about the PMS. He said it would take 5-8k up front, or an order for 50 units, to develop a PMS setup for our Bimmers.

I would be happy to try to work with him to set up a group buy arrangement if there were adequate interest on this and the other boards to warrant the development.

I might even be able to be convinced to put up some/all of the money to develop it, however I would certainly need some type of spiff or exclusivity to take the financial risk to do that...

Waddya think?

Steve

[97M3POWER]

Well as far as use, I can say one thing. There is no other Piggyback system like it.. 5k would be the equivalent of 5 units maybe 6. Should not be that big of a GROUP buy. :) I paid $975 for mine in my 1997 Eagle Talon TSi AWD *over 400 WHP to all 4 wheel with a 4 cylinder* The best investment that I have ever made.

I can say this. If Doug was to produce the PMS for our application, many of the "tuners/shops" that have purchased expensive equipment to burn eproms will surely feel the impact! I guarantee it! I will speak with him again about this also. Two people given him the once over might help! :D

[zenon]

UniChip check:)
SMT5 check:)
PMS:dunno I'm lost and this one's gonna be hard to find on google

[stimpee]

http://www.efisystems.com/

You're welcome!!!

Steve

[97M3POWER]

http://www.efisystems.com/

You're welcome!!!

Steve

Thanks stimpee .. and sorry zenon for not given the site information.

[paul e]

I just went to the PMS site, and read the installation and useage manuals for the Porshe.....Im a rank novice at this kind of tuning, and while I may write software for a living, my knowledge of electronics, and even simple electrical knowledge and practical experience is lacking....However, the instructions are so well done that I think even I could tackle this if they were to come out with a version for us...Even at $1200, like the Porshe unit, I think Id still be interested. Do you guys think that, if this were the ballpark price, it would place it beyond the 'avg' M3 forced induction home user interested in tuning his own parameters and only into the hands of the pro independant? Or do you think there are a number of home users who would be interested even if it were only for their own cars....If it takes $8k to get a run, and there are 8 interested users for a GB, then that might be enough?

I havent checked out too many alternatives yet...but for piggybacks, Shawn's system (PCIII) looks good, but definitely more complicated to wire up and use for guys like myself, and Unichip looks interesting, but there are a limited number of locations, and the ones Ive found are porshe shops?. SMT5 also looks interesting, but again, nothing seems as user friendly for electical numnuts like myself as the PMS.

[97M3POWER]

Originally posted by paul e
I just went to the PMS site, and read the installation and useage manuals for the Porshe.....Im a rank novice at this kind of tuning, and while I may write software for a living, my knowledge of electronics, and even simple electrical knowledge and practical experience is lacking....However, the instructions are so well done that I think even I could tackle this if they were to come out with a version for us...Even at $1200, like the Porshe unit, I think Id still be interested.

Exactly as I said, it is by far the best piggyback that I have worked with. Even easier then a S-AFC which we use on the Honda's and Mitsu for folks who don't want to take the $975 plunge. They all later regret it. By the time they purchase the S-AFC for fuel management and then the accompanying unit for Spark Advance, they are within a few hundred dollars of the PMS with the most advanced datalogging and tuning capability second to a standalone unit. Then again only till recently, Haltech's laptop tuning was done in a DOS only based application which was *UGLE* to say the least and hard to use for the NON-PRO. This compared to the point, click and enter of the PMS unit was a major point for a bunch of people down here to stick with the PMS piggyback system instead of gaining the extra HP with the standalone unit.



Do you guys think that, if this were the ballpark price, it would place it beyond the 'avg' M3 forced induction home user interested in tuning his own parameters and only into the hands of the pro independant? Or do you think there are a number of home users who would be interested even if it were only for their own cars...

If the kids in the import tuner market can afford the 400-600 dollars on a AFC setups with a car purchase price on avg $15,000-$20,000, I think a M3 owner could afford $500 bucks more with a purchase price of in excess of double! Don't you think?



If it takes $8k to get a run, and there are 8 interested users for a GB, then that might be enough?


I think it will depend on Doug on this one.. He is really busy with the current applications which his company supports. But trust me, as I said before, if this unit was to come onto the Bimmer market, it would surely eclipse many of the so called "tell me your mods and we will burn a chip for you that we have dyno tested" :rolleyes: guys out there. Not the Dyno Shops mind you, but the mail order eprom folks I think will get a bit of a beat'n!



I havent checked out too many alternatives yet...but for piggybacks, Shawn's system (PCIII) looks good, but definitely more complicated to wire up and use for guys like myself, and Unichip looks interesting, but there are a limited number of locations, and the ones Ive found are porshe shops?. SMT5 also looks interesting, but again, nothing seems as user friendly for electical numnuts like myself as the PMS.

I too think the PCIII that Shawn has did extensive research with is in my opinion the next best thing over the SMT5, UNI, Etc., but for a quick install and amazing datalogging package with 3 separate Setups all in one, PMS is hard to beat. IMHO