PD-01 Jet Engine

[racket]

Hi Patty

By restricting your NGV flow area the pressure ratio across the NGV will have to rise , if previously the NGV was "oversized " for the "restricted" flow rate caused by the turbine wheel then you have now restored a more balanced pressure drop across both NGV and turb wheel , turbine engines are about balancing out things in the most efficient way .

The slightly increased pressure ratio across your NGV now will be producing slightly higher gas velocities better able to increase power to the turbine inducer at the higher tip speeds .

Ideally you need to "clip/trim" your turb exducer back a bit to open up its flow area , with your current "lowish" temperatures your gas turbine engine doesn't need the amount of gas deflection (at the exducer) as the turbo did originally on the IC engine where the priority was extracting as much energy from the gases as possible , we only need to extract just enough to drive the compressor .

Theoretically if the exducer was opened up, and with an appropriate NGV angle and throat area , your BIG comp wheel won't be such a problem, its only a problem because of the "tight" exducer .

Currently your exducer tip angle is ~30 degrees ( 60 degrees from axially) , if it was "clipped" to produce a 40 degree angle ( 50 from axial) the flow area will be increased whilst still producing sufficient gas deflection to satisfy your comps power needs .

Have a look at the rather excessive clipping required for my 12/118 engine , postimg.org/image/4bmdneu9p/ , you would only need half that amount , something closer to what our TV94 turb exducers are jetandturbineowners.proboards.com/attachment/download/14

Cheers
John

[finiteparts]

Hi Patty!

Glad to hear that it is going the right way. It's nice to see that our gut feelings from the data appear to be right in line with the results that you are seeing...I have to commend you for working steadily to test out the concepts. I feel confident that you were on the right side of the peak efficiency lines (on the choke side, maybe not near choke, but on that side).

I apologize for the error in suggesting the exhaust nozzle concept as a means to move the compressor line back over the left side of the map...while "technically" the exhaust nozzle will move the op-line back over to the left side, it does this by robbing the available turbine expansion ratio.

Modifying the nozzle guide vane area is the correct way to move the op-line because it allows you to backpressure the compressor without robbing the available turbine expansion ratio. I think you are getting closer to having the required NGV effect area to match the turbines swallowing capacity.

Good job!

Chris

[racket]

Hi Patty

Your T2 at 153 C at 18 psi P2 - 2.22 PR represents a 122 C rise in compression , this is being achieved at a miserable 63.8% efficiency , it should be at least 10 points higher , something is still wrong, T2 should be ~135 C at that pressure ratio ...............is the temp pickup subjected to radiant heat from the flametube ?

Cheers
John

[madpatty]

Hi Racket,
Radiant heat is quite possible in my setup because there is no opaque thing between the proble and flametube as the diffuser design i am following is inverted type with open side towards turbine end.....
The pickup is inserted near the diffuser exit (where i used to have a P2 pickup earlier)....

Cheers,
Patty

[racket]

Hi Patty

Further to my comments about clipping your exducer ................have you actually measured the effective flow area between your exducer blades ??

If you haven't then I suggest you do it as a priority , you are going to have to work backwards from there to determine your flow and NGV throat areas.

Currently you feel the engine is running "better" because of the lower TOT , but its still trial and error testing and unless you also include thrust testing as well as fuel burn rates the "better" figures don't mean a lot .

Cheers
John

[madpatty]

Hi Racket,
Can you suggest me how to exactly measure the flow area at the exducer....what exactly is needed?

Patty

[racket]

Hi Patty

You need to take your caliper and position one side against the very tip of the exducer blade at its end, the other side of the caliper needs to be positioned across the gas flow ( at 90 degrees to flow) until it contacts the neighboring exducer blade at its closest point , which will be some distance "in " from the end on the blade .

The do the same at the root of the blade .

You'll end up with a "V" shaped throat , workout area and multiply by the number of throats

Cheers
John

[racket]

Hi Patty

I had a bit of a search through my stuff and found a Garrett turbo core with a 72mm inducer X 58mm exducer turbine wheel , hub was 22mm dia , 11 blades , passageway throats between exducer blading were ~10.5mm at the tip to ~3mm at root , giving an area of 1336 sq mms total for the 11 passageways , this is 2.07 sq inches - 0.01438 sq feet .

Using your latest data at 18 psi P2 and a 122deg C rise in compression , this will require a ~106 deg C drop through the turb , so a T I T of ~615 C - 888K , assuming 75% turb efficiency , there'll be a need for a 2.0 PR to produce sufficient power to drive the comp .

As you only have a 2.22 PR from the comp , say 2.1 PR going into the turb stage , theres very little energy left in the exhaust velocity .

Lets assume equal pressure drop in NGV and turb , so sq root 2 = 1.41 PR , this will produce gas velocities of ~1240 ft/sec .

Now density coming out of the exducer is at atmospheric static pressure and 509 C - 782 K , = 35.46 cubic ft/lb or 0.0282 lbs/cu ft .

Mass flow = speed X density X area = 1240 ft/sec X 0.0282 lbs/cu ft X 0.01438 sq ft = 0.5 lbs/sec = 30 lbs/min , or if you prefer metric 0.227 kgs/sec .

If we work on 0.5 lbs/sec at the NGV entrance and with a 1.41 PR across the NGV , then 2.1 divided by 1.41 gives us 1.49 PR at the throat at ~826 K , density 25.1 cu ft/lb - o ~0.04 lbs/cu ft .

Now back to our continuity equation , 0.5 lbs/sec = 1240ft/sec X 0.04 lbs/cubic foot X area , area = 0.01 sq ft or 1.45 sq ins or 936 sq millimetres , if we add on 10% for boundary layer etc , then an area ~1030 sq millimeters will do the job .

Cheers
John

[racket]

Hi Patty

When you were running T I Ts of 1,000 C your flows would have reduced to ~25 lbs/min - 0.41 lbs.sec - 0.189 kgs/sec , which might have been OK with a vaneless diffuser and a map width enhancing slot , but with a vaned diffuser you could have been running into flow instability which would have exacerbated your temperature problems .

One thing I can't understand is how your temperatures are so much lower now than previously , do you have any ideas ??

Cheers
John

[madpatty]

Hi Racket,
This is the question that has robbed my sleep for some nights now.

But even your calculations are showing that i am running in the best region of the map. 0.22 Kg/s is what i should expect at that PR as per the closest compressor map. Can't the answer be that simple that my turbine is able to flow this much?

Currently i am doing some tests by further blocking the NGVs and i want to share some data which i acquired by blocking 2 and then 3 NGVs :-

2 NGVs blocked

RPM   P2   T2   TOT



65000  7     99    527

75500  10   101  530

84600  13   116  527

87000  14   122  527



A4800  21   166  565

A6300  22   171  572

A7100  23   175  571

And when i tested with 3rd NGV blocked , the results were quite similar to the above:-

RPM     P2   T2       TOT



65600   7      93       535

76500   10    108      520


88400   14    131      517


A2600   20    167      543
A4700   21    174      549

A6300   21.5  175     550



What are the normal TOTs as one should expect while running the engine from startup till Higher PR which according to you are suitable for feeding into a freepower??

One thing more i wanted to ask how did you calculate the throat area of the turbine passage from the measurements 10.5mm(tip side) and 3mm(root side) AND moreover mine is a twelve blade turbine.

cheers,
Patty

[racket]

Hi Patty

After blocking a number of NGV passageways theres still no definite trend pattern developing , it would appear that the controlling influence is elsewhere .

You are able to flow your 0.22kgs/sec due to the lowish temperatures, if you put a freepower or jet nozzzle on the engine the temperatures will rise and your mass flow will decrease .

Despite you having 12 blades there'd be only a minor difference in flow area , these are only rough calculations .

With 10.5mm at tip and 3mm at root we simply break up the flow area into 3 segments , first is a 3 X 18 rectangle in the middle with 2 triangles of 3.75 by 18, one on each side , we add them together to get another rectangle of 3.75 X 18 , add on the 3 X 18 and we have a total of 6.75 X 18 = 121.5 , by 11 of = 1336 sq mms = 2.07 sq ins .

With 3 NGvs blocked , what is your current total NGV throat area ??

Cheers
John

[madpatty]

Hi Racket,
The current NGv throat area is about 940 sq.mm.

Cheers,
Patty

[racket]

Hi Patty

OK , so roughly what I calculated "theoretically " , your TOTs have been relatively stable for 14 psi P2 from all NGVs open to 3 closed , well within the range of normal error , 20 or so degrees isn't anything .

Theoretically the engine is running pretty well with those temperatures , its just the discrepancy between rpm and P2 that needs sorting , your T2 temperatures are giving the impression that your P2 gauge isn't reading correctly , if your P2 was higher then your T2 would be more in line with where it should be , also your flows through the turbine stage would be closer to whats expected , currently the available pressure ratio isn't enough to get that flow .

Its probably time to add a P4t pitot downstream of the turbine exducer to measure what is actually coming out of the wheel with regards to velocity , you only need a low pressure gauge as the maximum total pressure your engine will produce is probably 7 psit- 0.5 bar, but without a jet nozzle/freepower only a few psi of total pressure, probably handled by a big water manometer.

Ideally you need to have a length of "exhaust" pipe that the pitot can be installed in at least 300 - 500 mm downstream of the turb wheel so that the gases have time to form a "uniform??" flow before impacting the pitot inlet .

Cheers
John

[madpatty]

Hi Guys,
I have got a minor lubrication problem so please if anybody can suggest something...

Actually i have got two hydraulic pumps(external gear pumps)- smaller one for oil feed and larger one for oil suction.

But recently i have noticed that once oil heats up to say 70-80 degrees, the pressure gets reduced to 30 Psi whereas it is 70+ Psi when oil is cold....

Is that a problem in pump,
or i need a larger pump
Or such pumps are not good for hot oil(but i have seen people using such pumps here).

Cheers,
Patty

[finiteparts]

Patty,

That is normal. As the oil heats up, the viscosity drops and as a thinner fluid flows through the passages more easily, it relieves the pumps work input. For example, the kinematic viscosity of 15W40 at 0 C is about 1350mm^2/s, but when it is 80 C, it drops to around 20 mm^2/s. Also, in the bearing there is a high temperature and high shear (HTHS) condition that exists, which cause the fluid film that is supporting the rotor dynamic loads to be even thinner due to the local heating due to the HTHS. As the oil temp is increased beyond some temperature limit (Holset says 120 C), the ability of that film to support the bearing loads fails and metal to metal contact starts (see Stribeck curve for classic example)...this is why oil temperature is such a critical parameter to monitor. As the bearing clearances are the controlling flow area in the oil system, the warmer oil through the bearings flows through with less effort, ie less pump pressure required.

Additionally, a pump requires more power to pump warmer fluid (assuming constant viscosity), so let's imagine that the oils viscosity didn't change. As the oil temperature increased, the fixed pump power would be able to support less and less system pressure. Now, this is a very small factor since the oils viscosity changes so drastically, but it is worth mentioning just so your aware of it from a general pumping standpoint.

Finally, the bearing clearances change with temperature, so the controlling area of the oil system changes with temperature. The are a lot of different materials in the center housing, each with its own coefficient of thermal expansion. There are a lot of heat transfer paths, with the oil being the primary means of carrying heat out of the bearing housing....so there will be all kinds of temperature changes over the operational regime of the turbo which drive variation in bearing clearances. This one would be tough to even "though" experiment how the clearances change, but it should help to make the idea of controlling the clearances in the bearings as well as possible (ideally to the manufacturers specs). This is another minor effect, but worth being aware of.

I hope that helps!

~ Chris