Im still looking into the details of exactly how to control the transition from supercharger to turbocharger on my nova if I compound charge it like I am planning to do.

I have come up with what a I think is a really simple solution to the problem, and I would like to be shot down in flames if I have missed something obvious as far better to find out now then after having done it.


I am planning on configuring the engine to work like so in terms of where the air goes:

Air filter
Supercharger
Turbo
Intercooler
Intake manifold

Now the supercharger will be setup to flow about 300bhp of Air (hoping to gear it for approx 10psi), but I need the turbo to be able to flow 500bhp of air, so the obvious problem with that setup is that once the turbo spools it will be restricted in flow by the supercharger and worse still what will happen I believe is that the pipework between the turbo and the supercharger will be driven into vacuum by the turbo pulling through more air than the supercharger wants to flow, which means that the pressure ratio the turbo is actually going to be working at could be massive, as if I end up with say 15psi (30 absolute) after the turbo but only 10psi of vacuum (5 absolute) before the turbo, then its a pressure ratio of 6 for example.

So the problem that I need to solve is a way to allow the turbo to breathe more air than the supercharger can flow at high rpm.

Im thinking that the simplest way to do this is to allow a huge airleak between the supercharger and the turbocharger in order to allow the turbo to breathe extra air in that hasnt been through the supercharger.
I realise that while the leak is active it means that the supercharge will no longer be doing anything useful and will just be a parasitic loss on the crank with no benefit, but thats fine as once the turbo is spooled the thing goes like hell anyway.

The idea that I have come up with is to use a twinport wastegate connected to the intake pipework between the supercharger and the turbo, with the connection to the pipework being the opening that is side on to the wastegate piston.

The idea being that when the wastegate is shut the boost from the supercharger will not pass the piston.

Then to control the opening of the wastegate I was going to plumb the top port into the intake pipework between the supercharger and the turbo (so it will measure pre-turbo pressure) and the port that is on the bottom of the wastegate into the pipework between the turbo and the engine (so it will measure post-turbo pressure).

This means that the wastegate will then open when there is a big pressure differential (determined by the spring, probably a bar or so) between the pressure before and after the turbo, ie only when the turbo is spooled.

I realise that a manufacturer would never do it this way, the charger would need de clutching and they'd typically need to recirc the air due to being air mass based mapping rather than speed density etc, so Im well aware that you wouldnt do it this way if you were making 100,000 of the car and hence had a massive development budget for a specific control ECU and electronic valves instead.

Can anyone spot any flaws in my logic there?
As it seems such a simple solution that I cant beleive I havent ever seen anyone else doing it as obviously there are people far cleverer than me who have implemented compound charged solutions!

 

Makes sense to me as you have describe it,

Is this to purly reduce lag?
If it is would it not be better to twin turbo?


How did the old group b lancia s4's do it?

 

Just found this on google search dude
Dont know if youve read it or anything

A twincharging system combines a supercharger and turbocharger in a complementary arrangement, with the intent of one component's advantage compensating for the other component's disadvantage. There are two common types of twincharger systems: series and parallel.

Series

The series arrangement, the more common arrangement of twinchargers, is set up such that one compressor's (turbo or supercharger) output feeds the inlet of another. A sequentially-organized Roots type supercharger is connected to a medium- to large-sized turbocharger. The supercharger provides near-instant manifold pressure (eliminating turbo lag, which would otherwise result when the turbocharger is not up to its operating speed). Once the turbocharger has reached operating speed, the supercharger can either continue contributing pressurized air to the turbocharger inlet (yielding elevated intake pressures), or it can be bypassed and mechanically decoupled from the drivetrain via an electromagnetic clutch and bypass valve or one-way valve (increasing efficiency of the induction system).

Other series configurations exist where no bypass system is employed and both compressors are in continuous duty. As a result, compounded boost is always produced as the pressure ratios of the two compressors are multiplied, not added. In other words, if a supercharger which produced 10*psi (0.7*bar) (pressure ratio = 1.7) alone blew into a turbocharger which also produced 10psi alone, the resultant manifold pressure would be 27*psi (1.9*bar) (PR=2.8) rather than 20*psi (1.4*bar) (PR=2.3). This form of series twincharging allows for the production of boost pressures that would otherwise be unachievable with other compressor arrangements.

However, the efficiencies of the turbo and supercharger are also multiplied, and since the efficiency of the supercharger is often much lower than that of large turbochargers, this can lead to extremely high manifold temperatures unless very powerful charge cooling is employed. For example, if a Roots blower with an efficiency of 60% blew into a turbocharger with an efficiency of 70%, the overall compression efficiency would be only 42% -- at 2.8 pressure ratio as shown above and 20 °C (68*°F) ambient temperature, this would mean air exiting the turbocharger would be 263 °C (505*°F), which is enough to melt most rubber couplers and nearly enough to melt expensive silicone couplers. A large turbocharger producing 27*psi (1.9*bar) by itself, with an adiabatic efficiency around 70%, would only produce 166 °C (331*°F). Additionally, the energy cost to drive a supercharger is usually several horsepower, thus if it can either be disconnected electrically (using an electromagnetic clutch such as those used on the VW 1.4TSI or Toyota's 4A-GZE) or allowed to freewheel and vent to the atmosphere, several horsepower can be gained independent of the efficiency gain by switching to one compressor.

Thus, switching the supercharger off at a certain boost or RPM threshold is most desirable, since a large, inexpensive journal bearing turbocharger can be used which will provide more than enough pressure and flow at high RPM for most twincharged motors. However, a smooth switchover can be very difficult to accomplish for non-OEM twincharging applications.

Parallel

Parallel arrangements typically always require the use of a bypass or diverter valve to allow one or both compressors to feed the engine. If no valve were employed and both compressors were merely routed directly to the intake manifold, the supercharger would blow backwards through the turbocharger compressor rather than pressurize the intake manifold, as that would be the path of least resistance. Thus a diverter valve must be employed to vent turbocharger air until it has reached the pressure in the intake manifold. Complex or expensive electronic controls are usually necessary to ensure smooth power delivery.

 

So after reading that... What about making an aircon clutch fit the supercharger?? Switched bythe ecu output at a set rpm?

 

i might be misunderstanding this but the way ive understood it is that your wastegate connection from turbo to engine is bleeding off air back in before the turbo? or have i missed the point?

 

ignore that, your using it to open the wastegate

 

Rollinz, yes im very aware of all the possible different configurations of charger and turbo and which feeds the other or both side by side etc, its just the details of exactly how to simply control the switchover with readily available parts that is what its not so easy to google it would appear.

Not interested particuarly in things like declutching the charger as its just not going to offer any benefit really for my particular application as I dont care about fuel economy etc.

Just want to find a smooth and simple way to allow the turbo to breathe more air than the charger can supply when its spooled and for the charger to supply the engine with moderate boost when its not.

 

Yes I would have two signal pipes, one from before the turbo, one from after, then by putting them into the two different ports of the wastegate I would be effectively measuring the pressure differential between them and using that to open the wastegate, when it opens then it would allow air inwards to the inlet tract before the turbo (there may actually be a brief point where the inlet pressure is down to say 3psi of boost but the outlet is upto 18psi of boost where it might open very slightly too soon and allow charger boost to leak out for a few hundred engine rpm but I can live with that quite happily, once the turbo is spooled I basically dont care about the charger anymore until it unspools)

 

Sounds like it will work in theory mate... may have to be carefull with switch over points to ensure the switch takes place prior to any large inlet depression otherwise you are likely to have a very sudden and high overspeed condition on the turbos compressor / turbine.

There are some exhaust bypass valves knocking about that may be worth some investigation as an easy fit solution... yanks use them to bypass silencers electrically, wanted to fit one to the M5 but no damn room.

Additionally I would be interested to see if you could find a way to maintain some supercharger input into the compressor at all times to keep the efficiency high... that however, is likely to be a total mindfuck of a convertion requiring complex one way sealed valves, and the more I think about it, the less interested I would be in trying to do it. LOL

 

whilst complex in theory it is far more complex in practice

 

I understood it the second time i read it, but the first sounded like you were recirculating the air back in to the front of the turbo in some sort of crazy perpetual motion style!

 

I love this sort of thing, very very interesting!
I'll be interested to see the outcome.

I'd love to have an input, but this is way above my level of knowledge

 

Done this way I wont have any ECU control over it, it will be entirely mechanical.
But I think that the switchover should hopefully happen by the time inlet to the turbo sees even moderate vacuum.

I'll have to install a few boost gauges though so that I can watch it happen with the turbo set for moderate boost still to try and avoid that happening.

Might even set a boost cut at 5psi of vacuum for example so that if the inlet before the turbo sees that vacuum or greater (ie less absolute pressure) it simply turns off the solenoid control the turbo wastegate (the one in the exhaust) so stop the turbo trying to make big boost under that circumstance.


A weaker spring in the wastegate on the inlet would make it less likely to be a problem with the turbooverspeeding, but obviously it could end up fluttering like mad if its too weak.


For this project, I dont think that will be particuarly useful TBH mate, I am trying to only solve the problems I need to in order to get it to work "fairly well" rather than looking for a perfect solution which with my lack of budget and funds (ie I dont have a team of 100 and 50 million euros to spend) is never going to happen anyway.


Was racking my brains all last night trying to think why this wastegate idea WONT work, but cant come up with anything so far, which surprises me as if its that simple I just cant understand why its not quite commonly done!

 

It would be great if you could take some time to expand on that please Tony?

 

Havent read all the replies yet having to do some of that funny work stuff so don't have loads of time, I looked into it and did lots of research... I will pass on all my bookmarks links etc when I recover my hard drive if you like?

Basically, its not really compound charging if you use a switch over valve.... The problem with the switch over valve comes from the transition from blower to blower... It took VW A LONG time and lots of RnD to get this right, im sure you would live with a less perfect solution but its still meant to be a big task.

The waste gate idea has been used before, I will try ad dig out the guy who used it (think it was on a nissan in the states), this was the simplest solution. But again not perfect....


By far and away the best results that I have read about have come from true compound charging. Big turbo feeding a supercharger, no switch over valves etc. People have had gt35's coming on boost at the same kind of engine rpm as a gt28 with all the efficiency advantages of a big exhaust housing, and a compressor thats big enough to sit right in its peak efficiency zone for most of the time.

Have a look at this nutter

http://www.youtube.com/watch?v=DD2XAs-871o

And here is a compound charged evo

http://www.youtube.com/watch?v=hrLsnR8gNiA It doesn't really look 'fast' but it really is! I think the power delivery is so smooth its deceptive. He calls it twin charged but i remember reading about it and its compound charged, again no switch over.

 

Because im trying to make this work still on the standard ECU, controlling electronic devices isnt an option at this stage, but obviously if I fail to do it mechnically I can fit an autronic or something and then be able to control such things i the map instead.

 

The obvious disadvantage of a true compound set up comes from the multiplication of inefficiency. IMO a good charge cooler after the supercharger, and an intercooler on the turbo would work well. I say charge cooler after the supercharger as im sure space will be tight in this area.

 

Have you seen the compound charged evo 5 ?

It uses no clutches or bypass valves and works very well.

 

Chip, RPM and load switches are no problem. You can have a standalone switch if thats what you need. In fact I have most of the parts on my desk to make one if the need should arise.

Rob,

 

Rob DOHC, turbo going straight into the charger was my initial idea, and as you say there is no need for any control as such then, but there are other issues that I dont like, such as the amount of pressure the supercharger is exposed to in a system where the final boost delivered into the engine needs to be 40psi (absolute) or more.
Plus there are some SERIOUS heat management issues and for my particular application I really need to avoid masses of heat as I already have heat management issues due to the mid engine layout, thats the main reason I dont want a "true" compound charged setup.


Effectively on my setup it would be compound charged only during turbo spool.
So if you pulled away it would be:
supercharged only
then compound charged
then turbo only

Im fine with that, the turbo worked great at high rpm anyway on the previous engine, I just want something more before it spools and to help it spool a bit sooner to make the car more drivable at lower rpm.

 

You mean like the vid above? lol good find on the power graph though!

 

Jed, I have seen a few graphs etc posted but not any details of the implementation, the thing is though I dont actually want the supercharger doing anything at high rpm as it will push heat up so massively, im quite happy for it to just give me a helping hand at the bottom and then just do nothing useful at the top end, including it actually sapping a small amount of power to turn it.
Once the turbo is spooled, it does everything that I want from the engine anyway, its just what happens before its spooled that I want the charger for.

 

I see where your coming from, but I think the simplest solution would in practice be the most usable. There are superchargers that are happy with high pressure in the inlet. They use man seals Have a search around for that EVO5, he spent a lot of time researching the supercharger, specifically because of his worries regarding high intake pressure.

 

The problem with RPM switchining is that whilst it works great at full throttle on a dyno pull, it will have implications for when you are at that point in the rev range on a closed throttle and then open it etc.

If I need to do an electronically controlled solution then I will look into it, but this thread isnt about that, its just about someone finding a flaw in my logic for the mechanical solution.

There are hundreds of way to do this, most of them extremely complex and requiring lots of setting up time that will mean I dont have time to do it for years as I get so little free time, so what I want to know isnt what is the perfect solution in an idela world, I want to know if there is a reason that what I have come up with wont work well enough to make it worth doing?

 

If the turbo has a twin scroll exhaust housing why not get one of these ?

http://www.spracingonline.com/store/...ool_Valve/3643

It would be alot easier than compound charging.

 

You dont seem to understand my current view of the nova, Ive done the spend 10K and make an epic engine thing with it before, thats easy, what I want to do now is spend naff all, use bits I have already, and end up with something close for naff all money.

Doing this the way I am saying will cost me:
M90 eaton charger - 300 quid
Pipework - 100 quid
Wastegate - got some spare already



So once again, rather than coming up with alternative solutions, which I can do all day long as well, can you please spend some time thinking about why what I have come up with WONT work?


I am already aware there are better chargers, and better turbos and frankly better engines than a 200 quid snotty old saab engine, but I am trying to do this as cheaply as possible with bits I either already have or can get hold of very cheaply.


You need to think "very cheap way of it working well enough" rather than thinking "big budget ultimate solution" to understand where I now am with this car, my days of spunking thousands on it ended when I went to the saab lump, its a different viewpoint entirely now.

 

Two reasons:

1) From what Ive seen they dont actually make the sort of difference I want here, which is for the car to have torque at 2500rpm not at 5Krpm when the turbo spools currently.

2) I actually just really fancy having a go at compound charging anyway "just because" as I have never done it before


Not looking for alternatives, perhaps my initial post wasnt clear enough, I just want to know if there is a flaw in the way I want to do it?

 

Right I think I understand now.

I'm just going to throw this out there anyway.
Would it not be simpler, have the same effect, use less space, and potentially create less heat, to run a twin turbo setup?
Ben.

 

No it wouldnt, twin turbo control is actually quite difficult too, and then I would have all the added problems of the exhaust manifold design and fabrication etc too and then still wouldnt have torque from tickover like with a supercharger.

Plus you are completely missing the "just because I want to" reason for the supercharger and turbocharger specifically.

 

Lol, hairly muff.. Your mechanical solution has been implemented by people in the past and from what I could find out worked well enough... Give it a go an see

 

Good luck Chip - I will follow this with interest

I have limited experience of it, and it is very difficult indeed

 

Cool makes sense. What I was saying about an RPM/Load switch still stands. A magic box that looks at both rpm/load on a look up table controlling when the valve is opened. Not hard to do.

 

Well that sounds pretty promising then mate, lets hope that its a case of "great minds think alike" and not "fools seldom differ" then if others have come up with the same idea as I have to solve the problem.

I'll have to get googling for some decent examples when I have more free time

 

I did think of just adding a megasquirt or mtech or something to look at load and rpm and control some valves, but I think the mechanical solution is simpler and more elegant TBH, but time will tell if I am going to just end up wasting my time or not.
Good thing is that it will literally cost me next to nothing to try (providing I dont overspeed the turbo!), seeing as I'll be buying a charger anyway whatever system I use in the end, and I already have a spare wastegate lieing around.

Physically fitting the charger is proving a pain in the arse at the moment, but thats a different problem altogether of course, lol

 

Cool, good luck with it mate and let us know how it goes!

A whole megasquirt just controlling the valve would be a little overkill though lol. Im playing with VVT control box at the moment, hence the desk with useful switchy things on

And tbh I do like nice mechanical solutions, often easier to understand and manipulate, there aren't many reasons why a electronic controlled valve would offer huge benefits.

And good luck with finding space for it! I think once you have sorted the blower mounting etc, the rest will be easy to experiment with.

Rob,

 

Going back on track to the origional question chip..

How about using a closed loop boost controler to open/close the gate that you want to create the extra intake with?

 

A boost controller taking input from where and doing what with it mate?

The circumstances when I want the gate open are:
When the turbo is spooled

The mechanical system seems to cover that quite well, im not sure how an electronic boost controller would add anything?

 

To open the gate when needed if wired into throttle position will help control the transition ???

 

Im not sure that would really help TBH, you could only really STOP it opening with a boost controller, not open it at a time it wouldnt be anyway.

can always add one at a later date though if that seems like it would be useful, ie if its opening when I dont want it to etc.

 

Sounds crude but simple and potentially effective Chip, I would certainly give it a go 'just because' mate! lol