|
Post by finiteparts on Sept 1, 2014 14:28:34 GMT -5
Whoops! Sorry about that...I did grab the wrong map!
By the way...it looks like my last post got put up twice. I had a error message come up that said the post failed, so I reposted and now there are two. I tried to delete one and it deleted both. Weird...
~ Chris
|
|
|
Post by racket on Sept 1, 2014 17:40:19 GMT -5
Hi Anders
I've just checked a vid of the 10/98 engine at idle of 35,000 rpm and it had a T2 of 67 deg C, so temp rise of ~50 C degrees during compression , but I only had a TOT of ~550 deg C (higher of two thermocouples with only ~10 degrees difference) with a 89mm jet nozzle installed .
67 deg C is still a pretty high temp so some reduction in idling efficiency .
Have you checked your thermocouple and gauge accuracy ??
Cheers John
|
|
|
Post by Johansson on Sept 1, 2014 23:00:36 GMT -5
I got it now Chris, thanks for explaining.
John, I have tried another probe and gauge identical to the one I use now and they both show the same temp.
It is a bit strange this idling temp problem. I ran the engine with 6 bar oil pressure but that shouldn´t brake the rotor down once it gets up to idle, and the RPM sensor should restrict the incoming air more the higher the mass flow gets.
|
|
|
Post by Johansson on Sept 2, 2014 0:23:27 GMT -5
|
|
|
Post by racket on Sept 2, 2014 0:28:22 GMT -5
Hi Anders
6 bar shouldn't be a problem .
Could you please explain the exact shape and diameters etc of the duct immediately downstream of the turb exducer to where it enters the bend at the flange, over what sort of axial distance is the reduction in diameter .
I'm assuming the 4" bend has an ID of ~98mm.
I can't seem to reconcile having a P4t of ~8.5 psit in the pipe at your 3.5 PR 2.5 bar P2, this sorta represents a velocity of ~1600 ft/sec , density of gases at 725C is ~45 cu ft/lb , with a pipe crossection of ~0.081 sq ft thats a mass flow of ~2.88 lbs/sec , which we can't do with any efficiency , allowing a boundary layer reduction in the pipe from 98mm to say 90mm then the numbers do stack up as flow will be reduced to ~2.4 lbs/sec -145 lbs/min or pretty well right in the best part of the map at that PR , a T2 gauge should be reading ~187 deg C on a 15 C day .
Yep, get the T2 gauge running and we'll check the temps to see what sort of efficiencies we're producing , if they're all within the "best efficiency" ballpark for their PRs then flows must be within the "best" range as well , we then need to look at the TOT thermocouple accuracy or downstream restriction.............m m m m , now heres a thought , could there be significant swirl in the exhaust from the exducer at idling conditions creating excessive back pressure , any reduction in diameter downstream of the exducer will increase that swirl, maybe some straightening vanes required downstream mounted off a "bullet" hub fairing ..........just rambling here , racking my brain for a solution :-)
Just a matter of getting enough data to cross check against each other, sooner or latter something will "declare" itself as the reason for the temps.
Cheers John
|
|
|
Post by racket on Sept 2, 2014 0:33:31 GMT -5
Hi Anders
The windspeed gauge could also be moved around the entrance to get a rough average velocity
Cheers John
|
|
ashpowers
Veteran Member
Joined: February 2011
Posts: 207
|
Post by ashpowers on Sept 2, 2014 1:25:43 GMT -5
LOL @ Racket is "racking his brain"... ;-)
|
|
|
Post by Johansson on Sept 2, 2014 5:14:15 GMT -5
Hi Anders 6 bar shouldn't be a problem . Could you please explain the exact shape and diameters etc of the duct immediately downstream of the turb exducer to where it enters the bend at the flange, over what sort of axial distance is the reduction in diameter . I'm assuming the 4" bend has an ID of ~98mm. I can't seem to reconcile having a P4t of ~8.5 psit in the pipe at your 3.5 PR 2.5 bar P2, this sorta represents a velocity of ~1600 ft/sec , density of gases at 725C is ~45 cu ft/lb , with a pipe crossection of ~0.081 sq ft thats a mass flow of ~2.88 lbs/sec , which we can't do with any efficiency , allowing a boundary layer reduction in the pipe from 98mm to say 90mm then the numbers do stack up as flow will be reduced to ~2.4 lbs/sec -145 lbs/min or pretty well right in the best part of the map at that PR , a T2 gauge should be reading ~187 deg C on a 15 C day . Yep, get the T2 gauge running and we'll check the temps to see what sort of efficiencies we're producing , if they're all within the "best efficiency" ballpark for their PRs then flows must be within the "best" range as well , we then need to look at the TOT thermocouple accuracy or downstream restriction.............m m m m , now heres a thought , could there be significant swirl in the exhaust from the exducer at idling conditions creating excessive back pressure , any reduction in diameter downstream of the exducer will increase that swirl, maybe some straightening vanes required downstream mounted off a "bullet" hub fairing ..........just rambling here , racking my brain for a solution :-) Just a matter of getting enough data to cross check against each other, sooner or latter something will "declare" itself as the reason for the temps. Cheers John Hi John, The reduction to 100mm is cone shaped and starts 10mm downstream the turbine exducer. The restriction inside the bend can only be guessed but a 89mm jet nozzle at the end should produce a "stable" total restriction. No idea about any swirl, but I can add a set of straightening vanes if you think it would be worth trying out. Better take this in steps though to sort out what modfification it is that solves the problem. Soooo, the next test will see the following mods: x T2 reading x New seal between engine cover and turbine flange (to get rid of leak) x TIT reading (should confirm if the TOT gauge is correct or not) x 89mm jet nozzle fitted x Air flow measurement taken at air box entry (mass flow can ten be calculated) Anything else I should try or is it enough for one test? Cheers! /Anders
|
|
|
Post by Johansson on Sept 2, 2014 11:45:37 GMT -5
*deleted*
(some strange double post)
|
|
nersut
Veteran Member
Joined: September 2012
Posts: 223
|
Post by nersut on Sept 2, 2014 12:13:40 GMT -5
Hi Anders Good progress! Hope you sort the small issues soon, can't wait to see a videos of you, blasting down the road again! How is the outboard turbine build going? Very interesting too! Cheers Erik
|
|
|
Post by finiteparts on Sept 2, 2014 12:33:13 GMT -5
Hi Anders,
I think a wind speed gauge would work great...unless there is a high amount of velocity at the inlet that might lead to a high suction...we would be sad to see the engine FOD out after you decided to complete a windspeed meter ingestion test...Ha!
No, really I think it is a good idea because you don't need to calculate the flow velocity. Actually, you could prepare a chart showing the mass flow at standard atmospheric conditions based on the airbox inlet area and the inlet flow velocity, that way during the test, you could glance at the chart and see where your mass flow range is.
John, I think the potential for residual swirl in the exhaust is high. Usually there is only a small operating range where the exhaust from the turbine exits in a fully axial direction. The potential for flow issues in a curved tube with a swirling flow seem like it would be quite "loss-y" and thus there would be a very high static pressure component. Curved passages set up secondary flows due to the forces needed to "turn" the flow and there is usually a spec given for how soon you can turn the flow after a turbine. The "visual test" of Ander's pipe didn't seem like it would be an issue, but the data he took made it seem like there are flow issues. I am not sure how the residual swirl from the turbine would effect the secondary flows in the pipe, but usually, swirling flows tend to highlight areas of flow separation. The use of deswirl vanes would be great at one flow condition, but then as the residual swirl changes the deswirl vanes might be severely off-incidence and cause flow separation and blockage, leading to even higher static pressures at the turbine exit.
Turboshaft and power generation turbines are quite sensitive to their exhaust systems. Since they are not trying to use the residual kinetic energy in the exhaust stream for any purpose (thrust, a subsequent turbine stage, etc.) they try to diffuse the flow carefully in order to recover the kinetic energy into static pressure. It is common to hear that the exhaust side of the turbine is actually operating at a static pressure below atmospheric! This took me a little bit of time to come to grips with back when I first heard it. The beauty of this for a large power generation turbine (or any for that reason, just the big ones spend more money on natural gas!) is that the pressure ratio across the turbine is increased, so the required firing temperature is lower, thus lower fuel burn, etc.. Conversely, this also points the other side of this coin, increased losses in the exhaust lead to higher static pressures at the back side of the turbine, lowering the turbine pressure ratios and thus causing higher firing temps to make the required power for the system.
I struggled with a good way to describe this pressure effect, but I will give it a try and hopefully it will be clear. At the turbine exit plane, you have a certain total pressure available to "drive" the exhaust out through the exhaust plane farther downstream, made up of the static pressure component (potential energy component) and the dynamic pressure component that is due to the momentum of the flow, i.e. kinetic energy. If you generate flow losses, you are using some of that dynamic portion of the pressure to make the eddies, viscous shear, heat and other flow losses, thus using up some of that kinetic energy.
At the exhaust plane where the flow meets the atmosphere, the exhaust jet has to have the same static pressure as the atmosphere (known as the "free jet condition", unless the flow is sonic or above, then this doesn't apply). So if you have lost a portion of the dynamic pressure (the momentum driving the flow out), then you will need more of the static portion of the pressure to push the flow out. So with the static pressure fixed at the exhaust plane, due to the losses, nature readjusts the pressure balance so that you have a higher total/static pressure at the turbine than the case with no losses. But if you can recover a large portion the dynamic pressure into static pressure (kinetic energy into potential energy), then you don't need as much total/static pressure at the turbine. So in the ideal case, you can see that the static pressure at the turbine can be below the ambient pressure by as much as the dynamic component of the pressure at the turbine.
In fact, when you are assuming that the exhaust pressure equals ambient in your calculations, this is what you are assuming...perfect exhaust expansion (no losses) to ambient static pressure, which is not realistic, but commonly done.
Thus, even the exhaust system needs to be thought out quite a bit in order to have a well operating engine.
Good luck! ~ Chris
|
|
|
Post by Johansson on Sept 2, 2014 12:36:23 GMT -5
|
|
|
Post by finiteparts on Sept 2, 2014 12:44:13 GMT -5
Done! Hope you win! ~ Chris
|
|
nersut
Veteran Member
Joined: September 2012
Posts: 223
|
Post by nersut on Sept 2, 2014 13:03:49 GMT -5
Done, voted!
|
|
|
Post by madpatty on Sept 2, 2014 13:25:32 GMT -5
Done....hope you win Anders!!!
|
|