How to rig a twincharger up?
Discussion
I'm having my engine out for a rebuild and decompress-even-more. My mate has decided to do the same with his mini. To allow for a big intercooler, the blower (M45) is going in the passenger footwell, this will blow through an intercooler into a Metro turbo carb setup. After a bit of debating, it seemed a good idea to use the turbo as well since it was "just there". I like superchargers but I think with 10PSI, the bini blower uses about 28Hp. That's a lot at full tilt, so what if the blower was made to go offline?
Lukey is an air conditioning engineer on big stuff and likes his complexity and fancy pipework, I like simple ideas which work. There are 2 approaches:-
Mine:- Rig up intake->turbo->blower->intercooler->engine. Here the key resistance is the blower sucking through the intake turbine, however, gearing the blower to produce say 12PSI, will quickly allow the thing to spool up. The dP across the blower is proportional to the power it uses, so as the turbo starts working, the amount of power the blower saps will be minimal. (Question, do Eaton blowers like a warm charge?) I argue that the turbo could be set to produce more boost than the blower, the dP will be negative and so I can't see that the blower would offer too much resistance to it, it would just be passing compressed gas with a minimum dP.
His, Rig up the blower and turbo in parallel with a flappy valve (or 2) to allow the maximum boost component the free path. The snag is that if the turbo produces say 15psi and the blower is geared to produce 10, it will produce a dP even if it isn't in use and be fighting against a constant 10Psi.
What method would you use, assuming that space is cramped?
(I'm just having a blower, it's him who is having the twincharger.)
Stu
Lukey is an air conditioning engineer on big stuff and likes his complexity and fancy pipework, I like simple ideas which work. There are 2 approaches:-
Mine:- Rig up intake->turbo->blower->intercooler->engine. Here the key resistance is the blower sucking through the intake turbine, however, gearing the blower to produce say 12PSI, will quickly allow the thing to spool up. The dP across the blower is proportional to the power it uses, so as the turbo starts working, the amount of power the blower saps will be minimal. (Question, do Eaton blowers like a warm charge?) I argue that the turbo could be set to produce more boost than the blower, the dP will be negative and so I can't see that the blower would offer too much resistance to it, it would just be passing compressed gas with a minimum dP.
His, Rig up the blower and turbo in parallel with a flappy valve (or 2) to allow the maximum boost component the free path. The snag is that if the turbo produces say 15psi and the blower is geared to produce 10, it will produce a dP even if it isn't in use and be fighting against a constant 10Psi.
What method would you use, assuming that space is cramped?
(I'm just having a blower, it's him who is having the twincharger.)
Stu
Have thought about this in relation to my supercharged two-stroke diesel idea. While I'll be doing the initial development with just a supercharger, I want the final version to have both, precisely in order to have the supercharger drive power available to drive the bike instead.
Using a Rootes/Eaton blower I think your series setup is the most sensible. Ignoring losses, the power consumed by a Rootes blower is a function of the dP it is developing, and the dP is a function of the relative restrictions to airflow on the upstream and downstream sides. Once the turbo has spooled up to supply the boost the dP across the blower drops, ideally to zero if everything's matched right, and the power consumed by the blower drops to the minimum possible. Simple and effective.
I don't think it matters whether or not "Eaton blowers like a warm charge". The charge will only be warm once the turbo has spooled up, and in this state the blower's not doing any work anyway.
The only thing I haven't really looked into is whether it is better to have the supercharger or the turbocharger first. In my case the plan is to have a boost device of one type feeding a plenum, then two smaller boost devices of the other type leading from the plenum to the individual intake scrolls for the two cylinders, in order to minimise variations in cylinder filling caused by the unevenly-spaced intake pulses of a V-twin. Practical considerations of size, weight, drive belts and cost mean that this setup makes a lot more sense if the primary boost device is a Rootes blower and the two secondaries are small turbos.
I concur with your objections to your mate's plan. It doesn't reduce the dP across the blower to zero so it is a lot less effective at reducing the power consumed by the blower, and it's also a lot more complicated, so no point. It only really makes sense if you're using a supercharger with internal compression, and it still needs a clutch to disengage the supercharger drive once the turbo's spooled up to really make sense.
Using a Rootes/Eaton blower I think your series setup is the most sensible. Ignoring losses, the power consumed by a Rootes blower is a function of the dP it is developing, and the dP is a function of the relative restrictions to airflow on the upstream and downstream sides. Once the turbo has spooled up to supply the boost the dP across the blower drops, ideally to zero if everything's matched right, and the power consumed by the blower drops to the minimum possible. Simple and effective.
I don't think it matters whether or not "Eaton blowers like a warm charge". The charge will only be warm once the turbo has spooled up, and in this state the blower's not doing any work anyway.
The only thing I haven't really looked into is whether it is better to have the supercharger or the turbocharger first. In my case the plan is to have a boost device of one type feeding a plenum, then two smaller boost devices of the other type leading from the plenum to the individual intake scrolls for the two cylinders, in order to minimise variations in cylinder filling caused by the unevenly-spaced intake pulses of a V-twin. Practical considerations of size, weight, drive belts and cost mean that this setup makes a lot more sense if the primary boost device is a Rootes blower and the two secondaries are small turbos.
I concur with your objections to your mate's plan. It doesn't reduce the dP across the blower to zero so it is a lot less effective at reducing the power consumed by the blower, and it's also a lot more complicated, so no point. It only really makes sense if you're using a supercharger with internal compression, and it still needs a clutch to disengage the supercharger drive once the turbo's spooled up to really make sense.
love machine said:
What method would you use, assuming that space is cramped?
(I'm just having a blower, it's him who is having the twincharger.)
Stu
I'd buy a proper turbocharger, that will have excellent spool, as well as the ability to provide good power.
Turbo technology has moved on since the early 80's. If you cant find one good enough, then add a little anti-lag.
I think the series arrangement is the most sensible one, and if you're running them in series I think you want the turbo blowing through the supercharger. The supercharger will always produce some compression, the amount is defined by the displacement and gearing relative to the engine. What the compression starts from is the manifold pressure at the supercharger intake. If you have a throttle at the blower intake the pressure at the blower intake varies between nothing and 1 bar (absolute). If you have a turbo upstream of the throttle then the pressure varies between nothing and as high as you look.
The main reason I can see for compound charging is that the supercharger complements a very laggy turbo i.e. a turbo designed to run at very high boost levels. Normally, the turbo depends on its own boost to generate the exhaust gases to drive it. The more boost it produces the greater this dependancy. If you can generate those exhaust gasses immediately without waiting for the turbo to spool up then you cut the lag, of course. This only matters if the lag is significant though, otherwise the blower is just so much extra baggage to carry around.
If you're planning really high power levels then the inefficiency of the blower may become an issue, in that case you could use a pressure-activated bypass switched open by the pressure upstream of the throttle. If you're going to be running off load for any significant time you'd want a vacuum operated bypasss too - could be the same bypass with two actuators. The bypass doesn't need to be very big, it only needs to cope with the difference between the engine capacity and the blower capacity. For example on my 4.6 the bypass is around 30mm bore.
The main reason I can see for compound charging is that the supercharger complements a very laggy turbo i.e. a turbo designed to run at very high boost levels. Normally, the turbo depends on its own boost to generate the exhaust gases to drive it. The more boost it produces the greater this dependancy. If you can generate those exhaust gasses immediately without waiting for the turbo to spool up then you cut the lag, of course. This only matters if the lag is significant though, otherwise the blower is just so much extra baggage to carry around.
If you're planning really high power levels then the inefficiency of the blower may become an issue, in that case you could use a pressure-activated bypass switched open by the pressure upstream of the throttle. If you're going to be running off load for any significant time you'd want a vacuum operated bypasss too - could be the same bypass with two actuators. The bypass doesn't need to be very big, it only needs to cope with the difference between the engine capacity and the blower capacity. For example on my 4.6 the bypass is around 30mm bore.
stevieturbo said:
love machine said:
What method would you use, assuming that space is cramped?
(I'm just having a blower, it's him who is having the twincharger.)
Stu
I'd buy a proper turbocharger, that will have excellent spool, as well as the ability to provide good power.
Turbo technology has moved on since the early 80's. If you cant find one good enough, then add a little anti-lag.
couldn't agree more! get a well sized turbo that will make the power and spool quick! i would deffinately go true dual ball bearings on a small motor like yours!
also might want to look at increasing cc, better suited cam, going EFI as this allows a whole host of other goodies (like anti-lag!!
) and a well designed manifold. you could think about alky and wayer injection so you can up the timing a bit. this will get the turbogoing sooner. and maybe a small (like 25 shot max) shot of N2o to help get it spooled. N2O will also reduce intake temps no end!
thanks Chris.
If the only benefit of the supercharger is to ameliorate the lag from a big laggy turbo, then presumably it's only worthwhile if you're going to use a big, stonking laggy turbo.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
A thought has just occurred to me: the shape of the boost curve will be largely determined by the boost device at the atmosphere end. So if the supercharger precedes the turbo, you'll get a nice friendly supercharger-shaped boost characteristic, while if the turbo comes first you'll get a mental turbo-shaped one which requires extra jiggery-pokery to tame it.
cyberface said:
If the only benefit of the supercharger is to ameliorate the lag from a big laggy turbo, then presumably it's only worthwhile if you're going to use a big, stonking laggy turbo.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
Well the metro runs a garret T3 so yes in respect to a 1275 a-series its a big laggy turbo.
annodomini2 said:
cyberface said:
If the only benefit of the supercharger is to ameliorate the lag from a big laggy turbo, then presumably it's only worthwhile if you're going to use a big, stonking laggy turbo.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
Well the metro runs a garret T3 so yes in respect to a 1275 a-series its a big laggy turbo.
so why not look into a Audi 1.8T turbo??? they max out about 150-180bhp and spool intantly on a 1.8. also you may beale to get a rebuilt with a ball bearing core to help spool even more!!!!
thanks Chris.
Audi s3 or s4 use a pair of small turbo's i was looking at them the other weekend, might be idea for your job, sounds like an ideal case for a hybrid tbh.
Thinking about it I guess it was s4 as from memory that engine is porsche designed so more likely to have been the elegant designed twin set up i saw.
Thinking about it I guess it was s4 as from memory that engine is porsche designed so more likely to have been the elegant designed twin set up i saw.
annodomini2 said:
cyberface said:
If the only benefit of the supercharger is to ameliorate the lag from a big laggy turbo, then presumably it's only worthwhile if you're going to use a big, stonking laggy turbo.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
Well the metro runs a garret T3 so yes in respect to a 1275 a-series its a big laggy turbo.
A very small T3. Mine used a Cossie compressor, with a 0.36 turbine housing and clipped blade.
It was quite laggy, making full boost ( 20psi )around 3500-3800rpm, then pulled very hard to circa 7500rpm.
I could break a std A+ box pretty much at will ( and did with about 6 before going straight cut ), I broke one Tran-X straight cut gear, broke several gear casings ( although never to the extent of doing any damage, they just cracked around the main bearing ), and finding a clutch that gripped was extremely difficult.
Of course turbo technology, along with mappable ignition ( welded dizzy is crude ) and of course fuel injection, have all moved on, so next time around which will hopefully be next year, I will make a lot more power, with a better turbocharger, over a wider rpm range.
Everything about it then was crude, but on a very tight budget.
I just hope someone has finally made a decent clutch since the early 90's.
Pigeon said:
A thought has just occurred to me: the shape of the boost curve will be largely determined by the boost device at the atmosphere end. So if the supercharger precedes the turbo, you'll get a nice friendly supercharger-shaped boost characteristic, while if the turbo comes first you'll get a mental turbo-shaped one which requires extra jiggery-pokery to tame it.
This makes me think of two things;
1) that you need to worry a bit about the temperature of the charge coming out of the turbo as it will reduce the effective capacity of teh supercharger,
2) The torque curve will be shaped by whichever device has lowest capacity at any point in the power curve, so it'll probably end up with a characteristic similar to the supercharger, but with a bit more power up the top as you're doing the compression with the turbo rather than the supercharger (so have less crank losses).
mave said:
This makes me think of two things;
1) that you need to worry a bit about the temperature of the charge coming out of the turbo as it will reduce the effective capacity of teh supercharger,
2) The torque curve will be shaped by whichever device has lowest capacity at any point in the power curve, so it'll probably end up with a characteristic similar to the supercharger, but with a bit more power up the top as you're doing the compression with the turbo rather than the supercharger (so have less crank losses).
The supercharger is constant inlet volume flow rate but the turbo is changing the density at the inlet, you could think of it as the turbo increasing the effective capacity of the blower. You can also fit a recirc valve to unload the blower once the turbo takes over.
The installation sounds horribly complicated and unless you've got some very unusual requirements that can't be met by a turbo on its own, I think the only reason to do it is as an engineering exercise. Must say I really like the instant toe twitching throttle response of a big positive displacement supercharger, but apart from that the turbo route has a lot going for it.
stevieturbo said:
annodomini2 said:
cyberface said:
If the only benefit of the supercharger is to ameliorate the lag from a big laggy turbo, then presumably it's only worthwhile if you're going to use a big, stonking laggy turbo.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
In an A series Mini I'd have thought that big turbos (with corresponding monster torque) would eat gearboxes for breakfast?
How about 3 tiny turbos with high rev potential? Pipework may be a tad complicated though.
Well the metro runs a garret T3 so yes in respect to a 1275 a-series its a big laggy turbo.
A very small T3. Mine used a Cossie compressor, with a 0.36 turbine housing and clipped blade.
It was quite laggy, making full boost ( 20psi )around 3500-3800rpm, then pulled very hard to circa 7500rpm.
I could break a std A+ box pretty much at will ( and did with about 6 before going straight cut ), I broke one Tran-X straight cut gear, broke several gear casings ( although never to the extent of doing any damage, they just cracked around the main bearing ), and finding a clutch that gripped was extremely difficult.
Of course turbo technology, along with mappable ignition ( welded dizzy is crude ) and of course fuel injection, have all moved on, so next time around which will hopefully be next year, I will make a lot more power, with a better turbocharger, over a wider rpm range.
Everything about it then was crude, but on a very tight budget.
I just hope someone has finally made a decent clutch since the early 90's.
Did you use a rally plate and a double grey diaphragm? I plan to get my 6 paddle plate re-surfaced with whatever I can find and run a double grey (which is what I have now). I was wondering how a rally plate would fare with the double grey. It seems to be par for the course to use a road/rally plate on the road, regardless of spec. I'm going to be running straight cut gears but no turbo. As the engine is going to be out in a few days, if I found a K1100 head, I might be tempted to use that as well. Lukey is going to stick with the 5 port head and the metro turbo gear.
Actually, it was a re-con verto cover for a 1000cc mini and an Ap paddle plate lol
Go figure.
I tried 3 turbo clutches, including a genuine Unipart item, and about 3 other clutches ive no idea what they were.
None gripped except this one. It was a cheapy Balco recon unit, and obviously a fluke. I have since ofund about 2 other covers I beleive would work well too. Just the luck of the draw really. My pedal is extremely heavy compared to most normal clutches.
Ive had to get a new friction surface on it a couple of times so car as the paddle does eat into it and the flywheel.
That said, I drove it back when it was my only car, and probably covered some 60,000 miles while it was turbocharged. The fingers on the forks are actually very worn now, where the cup acts on it.
I was never going to revert to the old crappy clutch design, as apart from being old and crap ( and very hard on thrust bearings ) the arm would have interfered with my IC pipework.
Go figure.
I tried 3 turbo clutches, including a genuine Unipart item, and about 3 other clutches ive no idea what they were.
None gripped except this one. It was a cheapy Balco recon unit, and obviously a fluke. I have since ofund about 2 other covers I beleive would work well too. Just the luck of the draw really. My pedal is extremely heavy compared to most normal clutches.
Ive had to get a new friction surface on it a couple of times so car as the paddle does eat into it and the flywheel.
That said, I drove it back when it was my only car, and probably covered some 60,000 miles while it was turbocharged. The fingers on the forks are actually very worn now, where the cup acts on it.
I was never going to revert to the old crappy clutch design, as apart from being old and crap ( and very hard on thrust bearings ) the arm would have interfered with my IC pipework.
have you thopught about going for a GT28?? they are very quick spooling (thanks to proper dual ball bearings) and i dont think they cost the earth if you get the from the staets! and they are of much better design than the older T series (used on the cossy).
also with mappable ignition and injection (go on get a magasquit hehe) you can build in boost maps for the gfears and rpm, making it feel much like a NA engine once upto boost!
thanks Chris.
also with mappable ignition and injection (go on get a magasquit hehe) you can build in boost maps for the gfears and rpm, making it feel much like a NA engine once upto boost!
thanks Chris.
GreenV8S said:
The installation sounds horribly complicated and unless you've got some very unusual requirements that can't be met by a turbo on its own, I think the only reason to do it is as an engineering exercise. Must say I really like the instant toe twitching throttle response of a big positive displacement supercharger, but apart from that the turbo route has a lot going for it.
In my case, I do have unusual requirements
Got to have the positive displacement supercharger to provide scavenge air under all operating conditions. The turbo would simply increase the efficiency. Also, it's for a bike, so lack of lag is very important...
chuntington101 said:
have you thopught about going for a GT28?? they are very quick spooling (thanks to proper dual ball bearings) and i dont think they cost the earth if you get the from the staets! and they are of much better design than the older T series (used on the cossy).
also with mappable ignition and injection (go on get a magasquit hehe) you can build in boost maps for the gfears and rpm, making it feel much like a NA engine once upto boost!
thanks Chris.
also with mappable ignition and injection (go on get a magasquit hehe) you can build in boost maps for the gfears and rpm, making it feel much like a NA engine once upto boost!
thanks Chris.
I'll probably be restoring/upgrading it next year, so who knows what might happen
GreenV8S said:
mave said:
This makes me think of two things;
1) that you need to worry a bit about the temperature of the charge coming out of the turbo as it will reduce the effective capacity of teh supercharger,
2) The torque curve will be shaped by whichever device has lowest capacity at any point in the power curve, so it'll probably end up with a characteristic similar to the supercharger, but with a bit more power up the top as you're doing the compression with the turbo rather than the supercharger (so have less crank losses).
The supercharger is constant inlet volume flow rate but the turbo is changing the density at the inlet, you could think of it as the turbo increasing the effective capacity of the blower. You can also fit a recirc valve to unload the blower once the turbo takes over.
What I was thinking about, is that the supercharger is also constant exit flow volume (isn't it?). Turbochargers tend to heat the air more than superchargers, so for a given boost level at supercharger exit, the air temperature will be higher, so the density (and hence mass flow)lower.
mave said:
Turbochargers tend to heat the air more than superchargers, so for a given boost level at supercharger exit, the air temperature will be higher, so the density (and hence mass flow)lower.
I dont think that statement is true at all.
depends on turbocharger used, and supercharger.
Lets see a crappy old Rootes blower with no IC pump 20psi for example, compared to an efficient turbocharger at the same boost level.
In general, turbocharger compressors are a LOT more efficient than rootes superchargers.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff