Electric supercharger discussions

[alexanderfoti]

First off, this isnt going to be a discussion about something like this:



I picked up a secondary air injection pump from a 7 series bmw. It looks like this:



Using a boost pipe into a bottle cap gives me an easy way to measure the boost pressure:



Using my highly accurate gauge (ahem), I measured around 3-4psi I think.

Here the gauge is at rest:



Here is is with 12v applied to the motor:



and a video of the proccess.

http://youtu.be/3MrDLBKy9G4 (might want to turn your sound down).

The motor draws 14.5amps at open flow and 12 amps with a blocked outlet. Do you guys thing there is any milage going down this route?

[alexanderfoti]

Okay so 5 in hg is approximately 2.4 psi, so not much at all, 2 in parralel would do twice the flow and probably around 5psi. I would have to find the space for them as well, so I think this option is out

[alexanderfoti]

at 16v the motor produced 4.5psi.

at 24volts the motor produced 7.4psi then blew up.

hmm

at least it was cheap

[gilburton]

Could you not remove the motor and fit a higher revving one ??

[alexanderfoti]

Could you not remove the motor and fit a higher revving one ??

That was my next thought. I do wonder how fast the internals can spin before exploding. Not sure I want to take the chance.

Its all riveted together, if its a metal impeller I may do that.

[tappy]

The reason the amps reading drops at zero flow is because the compressor is aerodynamically stalled. At this condition it's not moving air, but it's probably not generating its ideal pressure rise either. You need to measure the pressure at a few different flow rates.

I accept that to know whether the idea is viable you need to know the pressure it can deliver, but what you've really measured is the stall pressure, which isn't the same thing.
To know if it's viable you also need to know the flow rate it can deliver.

The fact that it delivered higher pressures at higher voltage (speed) is to be expected - that's what c.f. compressors do, but without knowing the volume it was flowing it still doesn't tell you much.

[alexanderfoti]

The reason the amps reading drops at zero flow is because the compressor is aerodynamically stalled. At this condition it's not moving air, but it's probably not generating its ideal pressure rise either. You need to measure the pressure at a few different flow rates.

I accept that to know whether the idea is viable you need to know the pressure it can deliver, but what you've really measured is the stall pressure, which isn't the same thing.
To know if it's viable you also need to know the flow rate it can deliver.

The fact that it delivered higher pressures at higher voltage (speed) is to be expected - that's what c.f. compressors do, but without knowing the volume it was flowing it still doesn't tell you much.

Yes thats true. I did measure it whilst providing some flow etc, it had little pressure drop with a little bit of bleed off if that makes sense.

I havent been able to find the flow rates on any of these typres of pump. My engine uses 23-25 cfm at 3600 rpm, but I dont know how to measure the CFM of an air pump.

[tappy]

What's the internal diameter of the outlet pipe?

[alexanderfoti]

What's the internal diameter of the outlet pipe?

16mm

[tappy]

Generally, the output air from a centrifugal pump is slowed down to Mn0.2 to 0.3, to reduce flow losses.
After the modest pressure & temperature rise in the smog pump that might be around 0.2*350m/s = 70m/s.

An outlet diameter of 16mm gives a flow area of .0002sq.m, so giving a flow rate of .014 cubic metres / sec. That's 0.5cu ft /sec or about 30cfm.

If your engine is a 600cc, then at 100% volumetric efficiency it should swallow about 39cfm. Admittedly at 3600rpm it probably doesn't have 100% vol. efficiency, but then supercharging is meant to be helping that...

[tappy]

How much pressure do you think you're after?
I think you said you want about 3hp extra, and let's assume that driving the supercharger takes 0.5hp, so you need +3.5 at the crank on a 10hp engine.
That means you need 30% more air mass flow. If you use an intercooler that keeps the air at ambient temperature, that's 30% more pressure i.e. about 4.5psig. If you don't use an intercooler then you'd need need about 1.5 pressure ratio - about 7psig?

[alexanderfoti]

Generally, the output air from a centrifugal pump is slowed down to Mn0.2 to 0.3, to reduce flow losses.
After the modest pressure & temperature rise in the smog pump that might be around 0.2*350m/s = 70m/s.

An outlet diameter of 16mm gives a flow area of .0002sq.m, so giving a flow rate of .014 cubic metres / sec. That's 0.5cu ft /sec or about 30cfm.

If your engine is a 600cc, then at 100% volumetric efficiency it should swallow about 39cfm. Admittedly at 3600rpm it probably doesn't have 100% vol. efficiency, but then supercharging is meant to be helping that...

Mines a 400cc So the initial calculations look like it could work to some extent?

[alexanderfoti]

How much pressure do you think you're after?
I think you said you want about 3hp extra, and let's assume that driving the supercharger takes 0.5hp, so you need +3.5 at the crank on a 10hp engine.
That means you need 30% more air mass flow. If you use an intercooler that keeps the air at ambient temperature, that's 30% more pressure i.e. about 4.5psig. If you don't use an intercooler then you'd need need about 1.5 pressure ratio - about 7psig?

I was aiming for something like 4-5 psig.

The good thing about an electric system is I can easily wire it up, see if it gives power gains, then figure out generating extra power bit (probably add a belt driven 30 amp alternator)

[tappy]

I've just noticed something actually - that measurement of 16mm, did that come from the very end of the outlet pipe? From the step in diameter it looks like that's designed to have a hose going into it. What's the diameter of the pipe further in? Looks closer to 10mm? That would only give 30cfm x (10/16)^2 = 11.7cfm. Nowhere near enough - you'd need two of 'em, and I don't think that would work...

[alexanderfoti]

I've just noticed something actually - that measurement of 16mm, did that come from the very end of the outlet pipe? From the step in diameter it looks like that's designed to have a hose going into it. What's the diameter of the pipe further in? Looks closer to 10mm? That would only give 30cfm x (10/16)^2 = 11.7cfm. Nowhere near enough - you'd need two of 'em, and I don't think that would work...

Yes your right. It more like 12mm on in the inside of the ledge for the pipe attachment.

Oh well.

Edit: Thanks for help with the calcuations, I probably should have done that first.

Im trying to stop myself buying a z482 when I havent really got a need for it (the l100 clone is running well).

[tappy]

Well, to generate 5psig from a centrifugal compressor takes a perimiter speed of about 175m/s.
So a 60mm diameter impellor would need about 55krpm, a 90mm impellor would need about 37krpm, a 120mm impellor about 28krpm and so on.
I've said it before - this is vacuum cleaner territory.

At the pressure and volume flow rates you want you'll be pumping about .02Kg/s of air, and the pump will be absorbing a bit less than 1hp. Many vacuum cleaners are 1000 to 2000W (1.5 to 3hp) so again, this is vacuum cleaner territory.

There are lots of electric motors for model aircraft props that develop that sort of power, though I'm not sure what voltage they run on.

[alexanderfoti]

Well, to generate 5psig from a centrifugal compressor takes a perimiter speed of about 175m/s.
So a 60mm diameter impellor would need about 55krpm, a 90mm impellor would need about 37krpm, a 120mm impellor about 28krpm and so on.
I've said it before - this is vacuum cleaner territory.

At the pressure and volume flow rates you want you'll be pumping about .02Kg/s of air, and the pump will be absorbing a bit less than 1hp. Many vacuum cleaners are 1000 to 2000W (1.5 to 3hp) so again, this is vacuum cleaner territory.

There are lots of electric motors for model aircraft props that develop that sort of power, though I'm not sure what voltage they run on.

Maybe I should be looking at driving a vaccum cleaner impellor from a belt somehow?

[gilburton]

Go on crack open that credit card and buy Tims Honda. You know you want to

[alexanderfoti]

Go on crack open that credit card and buy Tims Honda. You know you want to

I would love to, but I absolutely loathe with a passion, the superdreams! Already been down the superdream route

I have ended up using my diesel bike a lot more than I thought I would, so the bike has to be something im prepared to ride around central london traffic.

A turbo'd Z482 in my bike with the gearbox is the ideal solution for me, and what I will probably end up working towards.

[gilburton]

Another option would be one of these but you'd have to ask about reliability??

http://www.pricepartmotorcycles.co.uk/o ... 13-hp.html

[alexanderfoti]

Another option would be one of these but you'd have to ask about reliability??

http://www.pricepartmotorcycles.co.uk/o ... 13-hp.html

True, and I would still have all the yanclone issues like:

1100rpm idle
vibrating the fillings loose at idle speeds (although it seems worse becuase I have a fairing and winsdcreen that flap around at idle)
direct injection relatively intolerant of veg oil (z482 is indirect).
Z482 is much quieter due to water cooling

And
£525 for this engine or £300 + watercooling bits, for the 482. Although its trivial to put the 458cc engine in, less so for the 482 as its not flat bottomed.

[gilburton]

Ah yes but it would be virtually a straight swap and you've got your extra 3 hp with no hassle

[alexanderfoti]

Ah yes but it would be virtually a straight swap and you've got your extra 3 hp with no hassle

that is true.

Has anybody every dynod them, 3hp from 50 cc seems quite large?

[gilburton]

It's nice to tinker but quite honestly I think you are on a bit of a wild goose chase.
Many engineers with better skills(and more money) than most of us have tried various ways of extracting more power from these singles with little success.
I'm sure if it was that easy it would have been done by now.
These engines don't respond to "go faster" mods because their basic construction/use is for constant revs and not the variable conditions in motorcycle use.
In my opinion your only realistic method would be to go for a larger engine of some sort where you will get nearer 15-20 hp.
It all depends on what type of engine you can get at a price you can afford??
As far as the larger single engine I believe he started fitting them to his Enfield conversions and he had some failures??
best thing is to send him a question and ask about failures and the extra power at the same time as he has had the practical experience

[alexanderfoti]

It's nice to tinker but quite honestly I think you are on a bit of a wild goose chase.
Many engineers with better skills(and more money) than most of us have tried various ways of extracting more power from these singles with little success.
I'm sure if it was that easy it would have been done by now.
These engines don't respond to "go faster" mods because their basic construction/use is for constant revs and not the variable conditions in motorcycle use.
In my opinion your only realistic method would be to go for a larger engine of some sort where you will get nearer 15-20 hp.
It all depends on what type of engine you can get at a price you can afford??
As far as the larger single engine I believe he started fitting them to his Enfield conversions and he had some failures??
best thing is to send him a question and ask about failures and the extra power at the same time as he has had the practical experience

Yup, I figured as much quite early on. Im sure a actual supercharger like an AMR300 or AMR500 would function quite well, but most end up going to a bigger engine, as you say its quite a bit easier.

Indeed, I agree that speaking with Henry is the best to figure out the reliabilty of the engines. Im sure they can be had cheaper from China and a slow shipping method. All food for thought.

[Blunt Eversmoke]

Ah yes but it would be virtually a straight swap and you've got your extra 3 hp with no hassle

that is true.

Has anybody every dynod them, 3hp from 50 cc seems quite large?

Why, swirl-chambers are usually more powerful than DI and ante-chambers are more powerful than swirl-chambers as far as we're talking atmospheric and non-commonrail, because IDI can spin faster. (The known trade-off is a somewhat higher consumption for both kinds of IDI). Also, you have more cylinders (one more in this case) which usually means more power per weight since the crank is less massive and thus somewhat lighter for a given displacement. All perfectly logical.

Oh, and ain't the Kubota a water-cooled engine? Also a big plus on the power side of things...

[Blunt Eversmoke]

Now, a little question:

What is the rotating speed of the centrifugal smog pump(at least as specified on the motor)?

[pietenpol2002]

I aquired a couple of Ametek central vacuum motors/blowers similar to the one in the link below to consider as superchargers for the Yanclone. They're a form of a regenerative 3 stage blower theoretically capable of approximately 6 lbs of boost on a Yanclone (but not taking into account the usual losses - thermal, volumetric, etc.). But, I wouldn't consider them realistic candidates for several reasons. Their bearings are designed for handling rotational loads only and with the motor removed, you need to accommodate bearings designed to talerate side loading. They need to be over-driven on the order of 6:1 resulting in speeds in excess of 21,000 RPM's. Some models run at 25,000. And finally, they're not designed to operate in the environs of the open road. Impellers are of very light weight aluminum construction. Having even light weight aluminum parts spinning at that RPM so close to body parts I've grown to like, doesn't foster confidence. And of course they can be had on your friendly international auction site for a fraction of the new cost reflected in the link.

http://www.alliedelec.com/search/produc ... #tab=specs