Turbo engines

[ivanhoew]

i.d give mcinnes 6/10 bell 3/10 and a graham bell 8/10.
mcinnes book was helpful and concise , but had a few bits on selecting turbo's and sizing compressors that I seem to remember actually made no sense ,and one ended up with a much smaller compressor than needed.

bells book I found full of absolute/ incorrect statements and testosterone laden 'this is the law' sort of stuff . not for me .he seemed to have no tolerance for anything that didn't stick to his opinion ,and presented that opinion as fact.

a graham bell , gets a little caught up in engine specific detail ,but has some interesting explorations of sides of tc that don't often get talked about .

all of course in my own opinionated opinion!

[jimbo39]

opinionated opinion

have you read Warner's book?

I have read a fair way through A G Bell's book and as you say he goes very very deep. Well good book though

[Tony]

Mc Innese's "Turbocharges" book was actually pretty good back in 1976, but a great deal has changed over the last 37 years, and much of it is not really relevant to either today's turbos or modern engine systems.
At the time, it was a goldmine of information, as there was really nothing else out there back then, and it was what started many of us old timers off on turbocharging in the late 60's and through the 70's.

Maximum Boost by Corky Bell, printed 1997 covers a fair bit of ground, but is also now becoming a bit dated.
The book has its faults, but is still well worth getting.

Forced Induction Performance by A Graham Bell published 2002 is the most recent major work on the topic that I am aware of, and also well worth getting.
It is quite possible there are more recent books now available, but I have not really been looking.

Both the C. Bell and G. Bell books are probably the ones to have.

[jimbo39]

On the subject of chargecoolers/intercoolers:

I've been thinking of something for quite some time but never got around to doing it (yet). Instead of using air to water coolers, why not use a refrigerant? Using an air conditioning setup as it would normally be used but having the cooling rad inside the chargecooler instead of inside the car. I don't see any reason why it wouldn't work, other than possible heat saturation, but if it was spec'd and setup correctly it would be quite something IMO.

I suppose someone has already thought of this or maybe even tried it to find it was useless, or maybe the gains outweigh the costs?

[tt911er]

Should have Air conditioning system powered by heat
When browsing compressors to get 600cfm less than 10psi boost I found that physical size of housing will start to play some role. Does anybody have an idea which brand/parts are able to squeeze that capability in a most compact size?
Definitely charge will be also cooled after turbos. May be ended up water/air coolers (running air conditioning cooler fluid in those )for size wise, also as an aircooled engine I would prefer to get turbos cooled as well. Probably not ever needed but heating would love to have warm water available instead air/air heat exchangers.

About the books. I have Street Turbocharging/ Mark Warner first edition 2006- It's alike Corky Bell's Maximum boost with couple of project vehicles and Turbocharging Performance Handbook/Jeff Hartman.

[jimbo39]

in my opinion Garret will be your best bet for size vs capabilities. Where do you live?

[Tony]

The compressor and exhaust housings are going to be fairly large anyway, not only to flow 600 CFM, but also because of the optimum lower a/r required at such a low pressure ratio.

The best water air cores available are supposed to be the very highly touted "Laminova" cores, If adapting a pair of standard Audi or Jag water air cores is not practical.
There is some guy on e-bay right now selling these (brand new surplus) at around a third of normal price.
An air/air core is always to be preferred if the installation space is available, as it allows closest to ambient one step cooling, but having a rear engine your choices are rather limited.

The only refrigeration system that I can ever see working really well might be direct air cycle refrigeration.
The idea here, is to generate some excessive boost pressure (practical with twincharging) inter cool that as best you can, then expand the air back down to final required boost pressure through an expander stage.

For best results the expander needs to be an efficient air motor such as a screw supercharger.
The shaft power generated by the air motor helps to drive the original supercharger to compress the air in the first place. You don't get it all back, but it's mechanical energy worth recovering.
The biggest effect would be the massive thermodynamic temperature drop produced across the air motor through Joule Thompson expansion.

It should be reasonably easy to get the induction air down to just above freezing in the highest ambient environment.
Well below ambient cooling would be available instantly on demand, and it could be sustained at a high level indefinitely.

Hey Doug, this might make a rather interesting secret weapon for the Bonneville guys in the forced induction classes, eh ?

[stef-1]

These guys seem to be using some sort of expansion system after the intercooler;

http://www.rtugroup.com/revolutionary-t ... technology

I have always thought that you could use the turbine part of a turbo as the expansion device after the intercooler. The shaft could be connected to a geartrain that then connects to the flywheel end of the crank so that power is given back to the crank.

If I recall, Lotus dabbled which either this or the supercharger version of expansion with their "Omnivore" research engine. There was a paper on it but it doesn't seem to be on their website now.

[stef-1]

Found it but not on the Lotus site:

http://papers.sae.org/2005-01-1853/

And a research project on the theoretical aspects of it:

https://dspace.lboro.ac.uk/dspace-jspui ... /2134/4830

BTW it was used on their "Nomad" engine not the "Omnivore" as I incorrectly recalled

[jimbo39]

thanks stef, that was an interesting read. 6.5% power increase is quite alot considering the model.