|
Post by finiteparts on Jan 25, 2019 23:30:13 GMT -5
Just playing with my cycle program and added the ability to create Temperature - Entropy plots. Also added cycle stations and the specific work in the compressor and the HP turbine. I need to verify my results with air standard properties, but I think it is very close. Also, I haven't incorporated the impact of combustion products on the hot gas Cp values yet, but I do have the ability to use the combustor FAR to modify the CP calculations. I just need to pull that into the matrix calcs.
The gray lines are scaled in atmospheres...i.e. the bottom line is for 14.696 psi (1 atm), the second line is 29.392 psi ( 2atm), etc.... The blue dashed line is set to the compressor stage discharge total pressure (Pt3) and the red dashed line is the NGV discharge total pressure (Pt41).
Enjoy
- Chris
|
|
|
Post by finiteparts on Jan 26, 2019 15:56:50 GMT -5
I finally had a chance to pull the Cummins HP841 apart and get us a better look at the turbine wheel. Here are a few shots of this nice little wheel. I did measurements and here are some of the major points of interest... - Turbine blade tip diameter = 5.230 inches - Turbine blade hub diameter = 3.363 inches - Turbine blade trailing edge thickness = 0.020 inches - Turbine tip blade exit angle ~ 74.2 deg from axial - Number of blades = 23 - Shaft diameter at bearing locations = 0.62855 inches - Shaft diameter at gear location = 0.4370 inches - Turbine geometric throat area ~ 4.97 in^2 I have not been able to locate any information on what material that it is cast from, but it is a very nice cast part. I did find online that the turbine makes about 7 ft*lbf of torque at around 50000 rpm, which after gearing down provides 192 ft*lbf of torque, which is roughly 66 hp to the engine. From this, we can see that the turbine operates at a relatively low tip speed ~ 1141 ft/s Enjoy! Chris
|
|
|
Post by racket on Jan 26, 2019 18:30:45 GMT -5
Hi Chris
What a beautiful little axial wheel , lotsa deflection there .
I'm interested in the "bearings" and gearing with respect to our other freepower setups , it appears to have a "normal??" turbo bushing and seal arrangement, with a helical gear at the output , could you please post a few more pics of that "end" .
When I constructed my bikes freepower I had its shaft running in a long floating "bush" and was always concerned about the helical gears "movement" as a result of the bearing clearances .
Cheers John
|
|
|
Post by finiteparts on Jan 26, 2019 20:06:13 GMT -5
John, Yeah, I was a little surprised about the hydrodynamic bearings and the gear, due to alignment concerns...but the bearing is tight, so the centerline deviation on the gear would be pretty small. I haven't busted out the mics and gauges to measure the bearing clearances, but when I do, I will post them. Here are some additional pictures: As you can see here, it uses two piston rings to seal...and one of the rings is has an interesting overlap to minimize the gap leakage.
I will try to get some detailed shots of the thrust bearing, the thrust bearing washers and the gear. - Chris
|
|
|
Post by racket on Jan 26, 2019 20:49:25 GMT -5
Hi Chris
Thank you for the extra pics , very enlightening to see how they've set things up .
The gear doesn't appear to have any keyway , just a "compression" hold ??
Fairly large helical angle for plenty of "overlap" between teeth .
Thanks again, these pics will be looked at many times over the coming years :-)
Cheers John
|
|
|
Post by finiteparts on Mar 30, 2019 19:01:38 GMT -5
I found an interesting item to add to my turbomachinery collection and the seller had no idea what they had, so I got i for cheap! If you haven't had the pleasure of reading old Hot Rod magazines from the 1960's, you might not know what this is. But I did read my dads old collection of Hot Rod magazines when I was a kid and I knew instantly what it was when I saw it. It is a Turbonique mono-propellant rocket powered turbo-supercharger! Unfortunately, the fuel is not something that you can run down to the hardware store and pick up, so it is for the time being just a wonderful piece of history. The idea on these was that you initiated the mono-propellant decomposition reaction with some O2 flow and a spark plug, which then could be turned off and the chamber pressure/power was regulated by the flow of the mono-propellant. The compressor is very crude and I think it lack any inducer angle because it had to mostly sit dormant while the car was driven normally. The turbine is a radial inflow impulse type and I will get some pictures when I disassemble it to clean it up. Enjoy! Chris
|
|
|
Post by pitciblackscotland on Mar 30, 2019 19:39:51 GMT -5
Hi Chris, Nice I have the technical manual from turbonique, haven't read threw it yet. Cheers, Mark. upload images
|
|
|
Post by racket on Mar 30, 2019 19:43:26 GMT -5
Hi Chris
Now that was a rare find.........nice :-)
Thanks for shareing
Cheers John
|
|
|
Post by finiteparts on Nov 16, 2019 16:08:36 GMT -5
Recent additions to my radial turbine collection. These things are huge!!!! I stumbled upon these two rotors on flea-bay and even though they are marked scrap, I needed them for my collection. The pictures on the listing didn't highlight how large they are and I was anxious to get one before I bought the second one in case the measurements given on the listing were wrong. Part number is 451-0645 Rev02. I have not been able to find their original application, but I am leaning towards MHI...maybe. Inspection shows a few nicks on the shaft and one blade that might be the cause of scrap tag, but other than that, they look brand new! I will check the shaft run-out and if ok, I would say that a quick blending of the nicks and a re-balance would make them usable. The hex measures 1.144 inches across the flats...the shaft is 1.125 inches at the bearing locations. The Inducer is approximately 6.7 inches in diameter and the exducer is 5.500 inches in diameter. Here is a side by side with a ST-50 rotor...holy smoke it is huge! Here is a comaprison to my TV94 turbine...
|
|
|
Post by finiteparts on Nov 16, 2019 16:16:13 GMT -5
I have a GTCP36 load compressor wheel that has the the left hand rotation to match this turbine and it is ball park sized. I will still look around, but one day maybe this will make a nice large engine rotor. The guy I got it from had already checked the material with his spectrometer and it is Inconel 713. Here is a shot comparing it to the large CAT turbo rotors that had been some of my largest turbos to date. Unfortunately the rotation is reverse between these so I can't "borrow" the compressor wheel. Another shot relative to the TV94 rotor that I used to think was pretty large. Enjoy! Chris
|
|
|
Post by finiteparts on Nov 16, 2019 19:09:31 GMT -5
I spent a little time recently looking at the GTX line of turbos to see if the higher available peak pressures of their Gen II compressors could allow me to reach a sonic condition at the exhaust nozzle and perhaps produce shock diamonds in an afterburner plume. The compressor maps are very interesting in their features, especially with the full length inducers (no splitter blades). I really think that the main take-away that anyone looking at these maps should have is the power of the recirculation slot technology to widen the compressor map. So, I started off with the GTX5533R size, just because the larger size and relative range of compressor sizes seemed at first pass a good starting point. Since my thesis was that the ability to run to higher pressure ratios would increase the available pressure ratio across the exhaust nozzle, I selected an initial design point around the 5.25 PR and around 150 lbm/min flow. Turbine flow chart... So how do we go about matching components from map data? Well, we know that we have to satisfy physics, thus we know we have to conserve energy and mass across a system. We determine the conservation of energy via the calculated work in each component. If we balance the compressor work put into the engine flow with the turbine work extracted from that same heated flow, plus all the losses, then we know that we have satisfied the conservation of energy at least to a first principles level. We can check our conservation of mass by matching the compressor inlet flow (W2) plus the fuel flow (Wf) with the turbine inlet flow (W40). If these flows, once corrected can coexist on the maps, then we have a possible engine operating condition. Ideally, we would have the rotor corrected speed matched on the compressor and turbine maps to be sure that the point is viable, but the turbine maps provided are not complete. Real turbine maps show the operating conditions along speed lines and the efficiencies below to increase the fidelity of the solution, but the maps provided by the OEMs usually are just peak lines that connect the peak PR of each speed line and sometimes they state the peak efficiency of the turbine system. So, to do this I wrote a code that solves for the required work at the user specified compressor operating point. Once the required work is known, the compressor discharge gas properties are solved. The program uses a fixed Cp through each of the stages, but to increase the solution accuracy, the averaged gas properties are used to find the Cp iteratively and thus the enthalpy through each stage is more accurately determined. The turbine inlet temperature is specified by the user and is typically the limited by the material capabilities of the turbine stage. With Tt40 specified, the requried heat addition can be found, again iteratively, thus solving for the fuel flow. With the turbine inlet conditions completely specified, the required temperature drop through the turbine can be found. With the temperature ratio known, the pressure ratio of the turbine stage can be solved for via the isentropic relations, since the inefficiencies are captured in the required temperature drop. The nozzle pressure ratio then falls out and the final check is to calculate the corrected flows to see if we are on the map. After playing around with this for several iterations, I come up with a realization that I probably should have known from looking at T-s diagrams. The low efficiencies of the turbocharger components (compressor ~ 70%, turbine ~ 74%) consume too much of the pressure in the system and even with a compressor PR of 5.25, the available nozzle pressure ratio is only 1.59 (critical conditions ~ PR 1.849). In order to get the nozzle pressure ratio up to just meeting the critical pressure ratio, the compressor efficiency needs to come up to 78% and the turbine efficiency needs to be around at least 76%. This all assumes a Tt40 = 1650 F and standard day inlet conditions. As John has pointed out, the larger compressor wheels are not well matched to the turbine for our purposes. The 88 mm compressor was the one that matched up well with a high pressure design point. The use of a larger compressor wheel would be challenged to flow with this turbine arrangement even at the largest scroll. At this design point the compressor is absorbing 368 hp. The turbine is creating 498 hp of which 130 hp is being dissipated as losses in the bearings, windage, tip leakage, etc.. The turbine spec work is in a pretty normal range of 54.9 Btu/lbm, so it is not being worked too hard. With an assumed 99.8% combustion efficiency it would still be sucking down 26.9 gal/hr of kerosene! So, is my dream dead? No, I think the potential to use a vaned diffuser to get the compressor stage efficiency up and similarly, a properly designed NGV section to increase the turbine stage efficiency are very possible. I may not be able to use the GTX turbos in as produced arrangement, but it may still be possible to get a choked exhaust nozzle. I think the key is the new forged, milled compressor wheels that allow higher rotational speeds (1920 ft/s) needed to get the stage pressures up. Couple them with a well designed vaned diffuser that allows higher pressure recovery and thus produce higher stage pressure ratios. The recirculation device must be kept if the map width is not to be drastically reduced, especially at the high end of the map. I think the potential is still there, but it will be a hard goal to get to. It must also be remembered that the nozzle pressure ratio must be above the isentropic critical nozzle pressure ratio because the AB system imposes a noticeable total pressure loss upstream of the nozzle itself. Hopefully, one day we can get some shock diamonds in the exhaust. - Chris
|
|
|
Post by racket on Nov 17, 2019 15:16:17 GMT -5
Hi Chris The dream is obtainable , Anders JU-02 jetandturbineowners.proboards.com/thread/734/building-ju-02-gas-producer should produce sufficient total pressure in its jetpipe , its comp wheel doesn't have the specific flow (lbs/sec/sq inch) of the GTX wheels , but those "lower?" airspeeds going into and through the wheel will probably add a tad to efficiency when combined with the "shallower" inducer blade angle, an 80% effic shouldn't be out of the question, I've been measuring better than 80% with the 12/118 engines comp at more modest PRs up to ~3.5:1 , but with a "horrible" 62 Trim wheel As for jetpipe pressures , my old Garrett TV84 with its 1980's technology was producing ~12 psit of total pressure in the jetpipe , and on a couple of occassions in 2000 I saw 14 and 15 psit on the jetpipe pitot ..........the jetpipe temps were a tad hot at the time so I didn't linger for long. Keep the dream alive :-) Cheers John
|
|
|
Post by finiteparts on Jan 4, 2020 13:13:44 GMT -5
Got a new addition to my automotive gas turbine related collection! In case you don't immediately recognize it, like I did when I saw it for sale, this is a tachometer and clock that would have been in the Chrysler Gas Turbine car. It appears to be new old stick and shows no signs of being installed. There was a time, when a car's tachometer needed to go to 60,000 rpm. These turbines whistled happily along, not coughing and sputtering like the anemic anchors (liberally called engines) that haunt cars today. Enjoy! Chris
|
|
|
Post by glennmichelle on Jan 16, 2020 12:43:25 GMT -5
Hello all, First, please redirect me should this be the wrong forum topic. I recently purchased an Airesearch Turbine used for pumping fuel or water and I am trying to identify the unit so I can find documentation on it. The data plate has the following information. So far this is all that I have found on the unit. It is complete and has all wiring and control panel in place with instrumentation.
Description:Pumping Unit, Centrifugal, Fuel Part: 380124-1-1 Series: 3 Model PUP70-26 Order:AF33-657-7892 Serial: P-16355
|
|
|
Post by finiteparts on Jul 9, 2020 20:11:26 GMT -5
I had a few minutes of free time to re-explore the idea of producing sonic conditions at the nozzle throat of a small engine. This time, I started by just toying around with my evaluation copy of GasTurb12. I have to reiterate what a wonderful program this is, especially for the novice. In the past, the eval copy was not time limited and all the features were enabled for the turbojet configuration. Now, the eval copy (Version13) is limited to a month or something, but you might be able to find the older version 11 or 12 somewhere online. I would also highly recommend the book "Propulsion and Power, An Exploration of Gas Turbine Performance Modeling" by Joachim Kurzke (the author of the GasTurb program) and Ian Halliwell...it's expensive, but a really useful book. Back to my pursuit of generating a supersonic flow stream in the afterburner of one of these small engines. As discussed previously, the low component efficiencies of standard turbocharger bits makes this too arduous of a task to be undertaken with the stock housings. This has led to a deeper dive into the design of both the compressor diffuser stage and the turbine nozzle guide vanes to explore what can be achieved to increase component level efficiencies. A major part of this task is also the fact that these components do not operate in isolation, but in fact are part of a system and as such they must operate within the capabilities of other components. As many turbomachinery system designers have discovered in the past, component matching is challenging at best...at it's worst, it can render an engine from fully operating. As I write my programs and discover some of these interactions, I will try to share with the group. In that spirit, I thought that I would share this plot, generated in GasTurb with basic parameters that fit within the range of our turbocharger based components. GasTurb allows you to make some really nice carpet plots of parametric variable sweeps as you can see here. In the first plot, it shows the impact of the turbine stage efficiency on the ability to generate sonic conditions at the exhaust nozzle throat. The low efficiencies of the standard scroll turbine housings limit the ability of the engine to produce the required overexpansion needed to generate the shock diamonds in the exhaust plume. It may be hard to see, but the colored bands within the plot are the nozzle discharge velocity in ft/s. The red band that covers the right hand side of the plot is the region where the nozzle throat has gone sonic. If you start at the bottom where the turbine isotropic efficiency is and follow the lines upward, you will easily see that if you have a higher turbine efficiency, the nozzle throat will choke at lower compressor pressure ratios. You can see that for a 72% turbine efficiency, that even going to a compressor pressure ratio of 5.4, you cannot get the nozzle to choke. Thus, the standard scroll housings just won't do for this purpose. You might wonder why I am saying this when many of the turbine maps state peak efficiency numbers around 79 and 80%? Well, it turns out that the peak efficiencies of turbine stages, with scroll housings, do not usually occur at the higher rpm conditions. Usually, they are tuned for mid-band and even worse, they are usually stating the value for a single housing in a group of several A/R ratios. The next carpet plot shows the required geometric area of an exhaust nozzle running with an AB exit temperature of around 2930 deg F. It shows the ability of the higher pressure compressor discharge to achieve higher gas internal energies, which leave more energy after the turbine to push the exhaust gas out of the exhaust. Now also, it should be apparent that with the Tt5 gas temperature being cooler as you get further to the left, the density of the gas is increased and thus it is easier to push a given mass through a fixed area. What isn't shown here is that the residual total pressure after the turbine (Pt5) is also increasing as you move to the left, further increasing the gas density. Finally, I though that I would also show the temperature-entropy and pressure-volume plots that GasTurb produces. It is really an exceptional program. I will discuss these later...but there is just so much that you can gain from looking at these diagrams. - Chris
|
|