wingers
Member
Joined: October 2013
Posts: 10
|
Post by wingers on Oct 13, 2013 15:34:18 GMT -5
Hi guys, I'm new to the forum and to the DIY gas turbine scene but I'm keen to create a GT based on an automotive turbocharger that I have spare. So far, I've been reading up on this as much as I can but finding specifics is a bit difficult it seems. My plan is to use kerosene as the fuel and (although its a bit small), a Garrett GT3076 turbo. I've created a basic combustion/flame tube design and I think the below fuel pump should be suitable, but I'm not quite sure what to do for fuel injection. What are the options that are generally used? A car injector or an open nozzle with PWM pump seem to be possible routes? What are the pros and cons? www.ebay.co.uk/itm/HIGH-FLOW-FUEL-PUMP-FITS-VARIOUS-CARS-UNIVERSAL-/160327897190?pt=UK_CarsParts_Vehicles_CarParts_SM&hash=item2554499066Also, my fuel pressure won't be very high (100psi) with the above pump and I'm a little nervous that it wont be enough. Especially since I'm looking to startup on kerosene directly, with no propane so I'm hoping to get as much advice as possible from those who have been here before. Wingers
|
|
Deleted
Joined: January 1970
Posts: 0
|
Post by Deleted on Oct 13, 2013 16:20:59 GMT -5
Hi Wingers, welcome How have you worked out the size of your flame tube etc, what is the inducer area of your turbo As for a fuel pump, you kind of on the right track, but i would buy a Bosch pump, the one you looking at is a copy and i have had differing results from ones like this, so i stick to only Bosch. You mite want to take a look at a 044Bosch pump, plenty on ebay. I know they a little bit more but should pay you back in results, they do come with a small built in suction filter but i would fit an extra inline one to protect your investment, you can control the pump with one of these www.ebay.co.uk/itm/12V-30A-DC-Motor-Speed-Control-PWM-HHO-RC-Controller/221039328668?rt=nc&_trksid=p2047675.m1851&_trkparms=aid%3D222002%26algo%3DSIC.FIT%26ao%3D1%26asc%3D18216%26meid%3D1962089264673904199%26pid%3D100005%26prg%3D8342%26rk%3D2%26rkt%3D5%26sd%3D170739987443%26or you could run with a fuel bypass system with a line "T" with a flow control vavle back to tank Fuel nozzle...... have a look here, heatingpartscenter.co.uk/prodtype.asp?strParents=3238&CAT_ID=2466&numRecordPosition=1you will need to work out how much fuel you are going to burn, which is all related to the area of the front of your turbo... If you phone the company ask for Nick explain what you are doing, he has help me and other in the uk Hope some or this helps
|
|
|
Post by racket on Oct 13, 2013 16:34:01 GMT -5
Hi Wingers
Could you please verify the wheel dimensions of your turbo , my data on the GT3076 has the turbine wheel a bit on the small side for a gas turbine conversion .
Cheers John
|
|
wingers
Member
Joined: October 2013
Posts: 10
|
Post by wingers on Oct 14, 2013 14:35:01 GMT -5
Thanks for the link - I'll have a look and probably give Nick a call.
If I go with PWM control I was planning to use an Arduino as I've already got one for another project so know how to program them etc. I could also up-scale it to include other aspects of control as well if I need to further down the road. I did look at the bosch pump to start with but the combination of higher pressure and lower cost pushed me towards the one I linked to. Have you tried that one or just a generic "replica". How tight am I on fuel pressure do you think? 100psi seemed to be a reasonable gain over 70psi for the Bosch pump...
Wingers
|
|
wingers
Member
Joined: October 2013
Posts: 10
|
Post by wingers on Oct 14, 2013 14:43:08 GMT -5
Hi Wingers Could you please verify the wheel dimensions of your turbo , my data on the GT3076 has the turbine wheel a bit on the small side for a gas turbine conversion . Cheers John The spec I've got is as follows: Comp inducer 57mm Comp exducer 76.2mm Turb inducer 60mm Turb exducer 55mm From what I've read, the compressor should be fine but the turbine is a bit small. Again, I'm not sure how bad that is though? The reason I was going for that turbo is that I have one in the garage that is virtually brand new... The flame tube I've worked out is: 143mm diameter 342mm length inside an outer housing of 181mm Primary holes: 27 off at diameter 6mm (I'll probably open up three holes to 8mm) Secondary holes: 8 off at diameter 9mm Tertiary holes: 8 off holes at diameter 14.25 So far nothing is bought or made yet (except the turbo) so this is all flexible. Wingers
|
|
Deleted
Joined: January 1970
Posts: 0
|
Post by Deleted on Oct 14, 2013 15:19:36 GMT -5
Thanks for the link - I'll have a look and probably give Nick a call. If I go with PWM control I was planning to use an Arduino as I've already got one for another project so know how to program them etc. I could also up-scale it to include other aspects of control as well if I need to further down the road. I did look at the bosch pump to start with but the combination of higher pressure and lower cost pushed me towards the one I linked to. Have you tried that one or just a generic "replica". How tight am I on fuel pressure do you think? 100psi seemed to be a reasonable gain over 70psi for the Bosch pump... Wingers Hi the bosch will give you more, dont beleave all they print about replica pumps.... and the bosch will run well at 18 volt, even at 12 volts i think you will get 8 bar 116 psi not to sure about electronic controlers, i would not use one
|
|
wingers
Member
Joined: October 2013
Posts: 10
|
Post by wingers on Oct 14, 2013 15:24:26 GMT -5
That's good to know about the Bosch. What do you mean about electronic controllers? Would you go for the "Tee" and needle valve as opposed to pump modulation?
Wingers
|
|
Deleted
Joined: January 1970
Posts: 0
|
Post by Deleted on Oct 14, 2013 15:57:03 GMT -5
Hi a "Tee" or PWM is fine, but Arduino i not sure about....... i would first make a working engine, get it running well with as simple control system as posible KISS ( Keep It Simple Stupid ) not a reflection on you :-).....
|
|
|
Post by racket on Oct 14, 2013 15:57:26 GMT -5
Hi Wingers
Yep , those are my numbers, thanks for that .
Do you know the turbine scroll A/R number ...........it should be cast into the housing near the rectangular inlet , either 1.06 , 0.82 or 0.63 , this will influence the swallowing capacity of the turbine stage .
The GT compressor wheels have reasonably wide maps so you might just get away with the "undersized" turbine stage , flow and thrust will be below potential for the sized comp being used , you'll probably be able to reduce the flametube diameter somewhat because of that , back to 110mm -120mm , leave length the same .
Once the engine is built, please don't install a jet nozzle on the exhaust until you verify that the restrictive turb stage isn't causing surge and/or over temperatures.
Cheers John
|
|
wingers
Member
Joined: October 2013
Posts: 10
|
Post by wingers on Oct 16, 2013 6:24:49 GMT -5
The exhaust housing is 0.82. Although I know it will have an effect on the flow rate, I've not really seen it included in any of the basic "hand-calc" flame tubing equations.
Wingers
|
|
|
Post by racket on Oct 16, 2013 23:56:31 GMT -5
Hi Wingers
According to the GT 3076 turbine map on the Garrett site the 0.82 A/R turb scroll allows a corrected mass flow of 23 lbs/min at a max efficiency of 72% , you should be able to run up to a 3.25 pressure ratio out of the compressor, at again a 72% effic, at ~125,000 rpm.
Lets assume a 5% pressure drop across the flametube so an ~3.1 PR going into the turbine stage , as its a GT turbo it should cope with a T I T of , lets start at 900 deg C
The 23 "corrected" lbs/min = actual flow times the square root of gas temp ( absolute) divided by 288, divided by the PR going in .
So 23 = actual X 2 divided by 3.1
23 X 3.1 div 2 = actual
Actual flow ~35 lbs/min .
Yep , it'll just get into the compressor map very close to the surge line :-)
3.25 PR at 72% gives an ~160 deg C rise in compression for a T2 of ~175 deg C , you'll be needing an ~140 deg C drop through the turbine stage for a TOT of ~760 deg C , and needing a pressure ratio across the turbine stage of ~2.06:1 , leaving a 1.5PR ( ~7 psi total pressure) in the jetpipe
There'll be a temp drop of ~94 degrees C across the jetnozzle for a velocity of ~1500 ft/sec , giving a thrust of ~27 lbs
If the turbine stage was of a more appropriate size, thrust could be as high as ~40 lbs with that compressor wheel .
Hope this helps :-)
Cheers John
|
|
|
Post by racket on Oct 17, 2013 3:20:22 GMT -5
Hi Wingers
Jet nozzle diameter will probably need to be ~1.85" - 47mm diameter , but actual final size will be dependent on a number of variables so play it safe and fire the engine up initially without a jet nozzle attached , if the TOT temperatures are down around 500 C at full rpm then it should be safe to fit a jet nozzle to raise temperatures, I wouldn't go directly to the 1.85"-47mm but to a slightly larger diameter initially , maybe 2"-51mm to test if temperatures are OK and the engine is operating without surge, you'll be very close to the surge line , so discretion is required if you want the engine to survive as well as perform .
Cheers John
|
|
wingers
Member
Joined: October 2013
Posts: 10
|
Post by wingers on Oct 17, 2013 14:06:17 GMT -5
OK thanks Racket. I understand this will be a compromised engine but that doesn't bother me too much at this point. I'm more interested in getting something that works, and isn't going to damage the turbo (I'd like to re-use it in a car at a later date). It sounds like I'm ok to progress this project on that basis and get introduced to GT engines Once I'm up and running then I might keep an eye out for a more appropriate turbo and develop a mkII GT Wingers
|
|
|
Post by racket on Oct 17, 2013 16:33:20 GMT -5
Hi Wingers
The turbo should work OK as long as you don't put too much restriction( jet nozzle) on the outlet, it'll certainly allow you to gain some "turbine time" before making another one .
Cheers John
|
|
wingers
Member
Joined: October 2013
Posts: 10
|
Post by wingers on Oct 26, 2013 15:06:08 GMT -5
I'm about to start cutting metal for the combustor and am having second thoughts about the design. The jetspecs design is known to be reliable/resistant to poorly built parts but also on the large size. I've also read a lot of other threads and the general advice seems to be that shorter/smaller combustors should work fine so I'm thinking that I should maybe reduce the diameter/length a little to allow for some of the "generous dimensions". Certainly if you consider HRR based designs then the combustor is much too big. What are peoples experiences with using a small HRR based design for a kerosene combustor?
Wingers
|
|