Active Supercharger Control

[Gollum]

So I had the idea a few weeks ago, of trying to get the on/off economy nature of a turbocharger with the temperature and package size benefits of a centrifugal supercharger. If your EFI system is MAP based, I don't see why you couldn't use an electronically controlled wastegate of sorts to control the manifold pressure relative to throttle/load of the engine.

 

Everything I've found on the topic via google searching comes up with people who want to OVER spin the S/C and use a wastegate to bleed of excess air on the top. That's not what I'm interested in. I'm talking about designing a closed system that's matched for "heavy load" conditions, then add an air bypass system for light throttle/low load scenarios to allow for economical tuning possibilities.

 

I've always said that the biggest benefit a turbo has that I can really grab hold of, is that it turns off and that "boost lag" is a good thing allowing very predictable low RPM use. Tell your girlfriend to keep it under 3,000 and she can drive your 500hp car to the grocery store (if you're so brave), while a S/C engine doesn't always offer the same leniency. We've also all seen how factory turbocharged cars usually get almost identical highway mileage as their naturally aspirated cousins, while a supercharged variation of an engine just plummets in MPG.

 

So tell me the pitfalls. Help me see the obstacles I might not have noticed. I think I big one is that the wastegate vent would be pulling air into the engine at vacuum, while any air filter or such would be a serious flow restriction when bleeding off boost at cruise. If not properly designed the system might just sit at 0 vacuum/boost until the wastegate shut off turning the boost on full.

[roger280zx]

Or use a Mercedes style clutch to (basically) turn the supercharger on and off. They have been doing it for years and it really helps the economy/longevity of the engine. A few guys have adapted the M45 clutches to M62 and M90 eatons. I'm sure you could use it on a centrifigul; though, if thats what floats your boat.

[rusty260]

MR2 AW11 have that function,

i got one and supercharger are not "working" at lower load

can get decent milage.

 

good explanation here

http://mr2.org.nz/phpbb3/viewtopic.php?f=12&t=7552

[Gollum]

Thanks for the input guys. I'm not sure why I never thought about a clutch system, since a S/C IS a pure linear speed based device.

 

I'm not sure I like how the MR2 is vacuum based though, as even stated in that thread it'll turn on at higher freeway speeds, when in reality the engine load isn't very high. I don't see why you couldn't make a table with one axis being vacuum and one axis being throttle. So that way it should be fairly "load" based, with some programmable definitions to allow a "highway" range that leans out nicely.

 

I'm not tied to the centrifugal idea, it just seemed easier for a wastegate idea, since it's literally half of a turbo. I've always been more of a fan of a twin screw supercharger. I do wonder how much a huge twin screw supercharger will allow to flow when it's not spinning very fast though. Any thoughts on this anyone? Do you think it could/would be a flow restriction worth noting for economy sakes?

[Tony D]

 

[Leon]

The S/C Toyota Previa had an electromagnetic clutch that engaged the supercharger depending on load. Too bad the head or head-gasket cracked when ours approached 200k!

[Leon]

The supercharger is engaged on-demand by an electromagnetic clutch, based on input from the engine management system computer (the Engine Control Unit, or ECU).

 

http://partners.academic.ru/dic.nsf/enwiki/621088

[S130Z]

Why not use a clutch setup found on an A/C compressor?

[Gollum]

 

 

 

That was only a LITTLE random Tony. Still loved it though.

 

The S/C Toyota Previa had an electromagnetic clutch that engaged the supercharger depending on load. Too bad the head or head-gasket cracked when ours approached 200k!

 

 

Thank you for the info!!! Will definitely be looking into it. The Merc stuff I've seen is outrageously expensive for old used parts.

 

Why not use a clutch setup found on an A/C compressor?

 

I guess that'd be fine as long as you could convert it to a better pulley. You don't want to run 60%+ more power than NA on a standard V-Belt. I wouldn't even shoot for 30% actually.

[Gollum]

Now this is a better price!

 

http://cgi.ebay.com/ebaymotors/TOYOTA-PREVIA-91-97-2TZFZE-OEM-SUPERCHARGER-_W0QQcmdZViewItemQQhashZitem2eafa98842QQitemZ200515618882QQptZMotorsQ5fCarQ5fTruckQ5fPartsQ5fAccessories#ht_1678wt_1165

[Leon]

No problem, I always thought it was a cool setup. I never thought of applying it elsewhere though, that would be great to see! That van was an awesome sleeper.

[Tony D]

Don't forget, at the time the Latham Axial-Flow Supercharger was being marketed, it was billed as the only TRULY 'switchable' supercharger.

 

Since it had axial flow design, like a jet engine with rotor and stator pairs, when it wasn't spinning it let air pass through end-to-end with no restriction to the N/A operation of the engine.

 

Unlike most of the 'parasitic roots' without a bypass valve.

 

I knew a guy with one, (a Latham) and I've half a mind to buy it off him just to replicate it. The design was intriguing, and it's modular assembly technique made it suited to may different sized engines. Just pack on more stages for more pressure and flow!

[Gollum]

Well tony, you HAVE seen that there's a company continuing the Latham design haven't you?

 

Quick google came up with:

 

http://www.axialflow.com/products.htm

 

definitely looks interesting, but it also looks kinda "mom and pop", which isn't always bad. But I have no idea of the capabilities versus price.

[Tony D]

Oh dude, you just saved me $1000+ on acquisition and reverse-engineering costs!

 

/Peggy Hill Voice/ "I did not know that!"

[Gollum]

After reading WAY too much useless information, it seems like the guy running that company has tried to engineer a kit for several cars with little success. I think the problem is that the kit costs ends up being too high to be competitive for what the performance is on paper. He also doesn't seem to want to sell them bare, as he doesn't want to be "at fault" for peopling blowing up their engines with bad tunes.

 

I really wish he'd just sell them bare and work on building them in quantity, then let others design kits for their market of choice and sell them.

[Tony D]

One other thing to mention is I have discussed the use of a PID Loop Controlled Blowoff on the intake piping of ANY supercharger (mainly turbocharger) system to allow for no-boost lost shifts and anti-surge control on turbos where the compressor is sized for a maximum horsepower level, but will surge at lower engine speeds. The answer is to keep the minimum stable flow across the turbo compressor at all times. This separates hot-side wastegate control into something more closely related to maintaining a set turbine speed for maximum efficiency, and then blowing off excess pressure to remain at stable flow rates and aways from the surge line.

 

The PID loop could have tie-ins with the TPS for rate of change and active closure (sort of a derivative action) to premptively close off the overboard dump in anticupation of rising engine demand as the reciprocating portion of the engine comes on to peack efficiecny levels and demands all the flow the turbo is capable of producing.

 

It's in the archives...

[Gollum]

One other thing to mention is I have discussed the use of a PID Loop Controlled Blowoff on the intake piping of ANY supercharger (mainly turbocharger) system to allow for no-boost lost shifts and anti-surge control on turbos where the compressor is sized for a maximum horsepower level, but will surge at lower engine speeds. The answer is to keep the minimum stable flow across the turbo compressor at all times. This separates hot-side wastegate control into something more closely related to maintaining a set turbine speed for maximum efficiency, and then blowing off excess pressure to remain at stable flow rates and aways from the surge line.

 

The PID loop could have tie-ins with the TPS for rate of change and active closure (sort of a derivative action) to premptively close off the overboard dump in anticupation of rising engine demand as the reciprocating portion of the engine comes on to peack efficiecny levels and demands all the flow the turbo is capable of producing.

 

It's in the archives...

 

Actually yes, I've heard (or read rather) you mention it before, and the idea DOES intrigue me. The irony is when I mentioned the idea to my friend (who's a respected subie tuner), he kinda laughed and said "I have a manual and electric boost controlled setup because I have TOO MUCH boost at low engine speed. I don't need MORE!!!" I find myself quite envious of his "problem" but it definitely shows that a good modern turbo works fine in street application.

 

Where I think this idea works best, is in all out racing. And there's nothing wrong with that.

[yellowoctupus]

Just out of curiosity, why not use Eaton's existing design, where you have the bypass valve (internal or external)? Parasitic losses are pretty low on the M62's when you're just spinning the rotors producing little to no pressure.

 

As well, you wouldn't have a need for any TPS PID loops etc, you can run directly off manifold vacuum. High vacuum (idling or low load = open bypass valve, low vacuum/boost/high load = closed valve).

 

If you're not using an Eaton SC, you can still use one of their external bypass setups, I'm pretty sure the Ford 3.8's were external, the GM 3.8's were all internal as far as I can tell.

 

http://www.not2fast.com/turbo/maps/M62power.gif This is the power consumption graph on the older M62's, (3rd gen I think) As you can see, even if you're running a 2:1 pulley ratio, at 3,000 engine rpm WITH 5psi it's 7hp. With no boost, it's going to be even less than that. (besides, by 3,000 I think you'd want something to be going on).

[Gollum]

Actually reading up on the eaton, some of the newer ones are solenoid actuated, which in theory could be managed by the ECU instead of a basic vacuum actuator.

 

But do remember, even 7hp in loss is a ton for cruising. It only takes roughly 8-10hp to move the average small civic at 55mph. Now, aero losses are exponential, so at 80mph that's a while different ball game. But still, I'm thinking that ideally I'd like to see losses drop to lower than 1hp, much like how an AC pump that's off shows almost unmeasurable loss in power.

 

But assuming that graph is under load, with the supercharger on, I'd guess that at around 2,000rpm you're only loosing 1-1.5hp, which isn't bad. If the losses really drop that much off load, then the eaton sure looks like the best way to go about this on a budget, since it's already ready. So does anyone know of a twin screw with a designed in bypass?

 

EDIT: Nevermind. I'm an idiot. It makes sense now, that any positive displacement supercharge HAS to have a bypass.

 

So really I just want to control the bypass via ECU, which has probably already been pioneered in the aftermarket.

 

So yea, umm.... case closed? Unless we want to further discuss a pulley clutch to accompany the bypass activation. But that does beg the question, if you had a clutch on the pulley, couldn't you get away without needing the bypass valve? Assuming air flowed through the supercharger I'd assume this to be true, but with a roots, and even a twin screw I don't think air will move easily on it's own.

 

It makes me wonder how much those Mercedes really gain by having a clutch on their supercharger...

Edited September 9, 2010 by Gollum

[Tony D]

If the guy has too much boost at low rpms, he doesn't have proper boost control OR wastegate control.