upper / lower cowl pressures

Hi

We are continuing with refinement of our cowl air inlets after getting the bulk of the cooling issues sorted. I now have instrumentation for upper (plenum in our case) vs lower exit pressures and have got graphs for the various air speeds.

I've tried three different inlets (sharp, rough foam hemisphere and profiled Kuchemann shapes). A major part of my problem now is not knowing what a 'standard' VANs soft baffle configuration is to start with (it probably doesn't matter that much about TW vs Nose or specific model 6-9). Has anyone done this measurement at all? Normally it would be via an ASI and two tubes with picollo tubes. (anybody keen to? )

As an example of some of the data I've been working with - this was the Kuchemann profiles on a short run yesterday.



Thanks for any other data,

Carl

Interesting data. I don't have a nice curve like you do, but several years ago I built a simple manometer with valving so that I could measure pressure above and below the engine. I could measure pressure above vs static, pressure below vs static and pressure differential top/bottom. Static was simply cabin static. I don't have the data in front of me, but the pressure above the engine was very close to calculated maximum possible for the conditions. I have the standard cowl. My conclusion was that, while I may have extra drag, I could extract no extra cooling by changing anything on the top side of the engine. The bottom was another thing. The lower cowl had significant pressure as compared to static.

My hangar mate Pete Howell and I plan to hook up the manometer again and rig up some under cowl video with tufting to figure out how to get the air out of the bottom. Previous tufting pictures from the exterior of the cooling exit area has shown some significant problems there. Somewhere on these forums I posted pictures of that work.

Can you tell us how you collected the data please? I see what looks like data series, and I also see what looks like clouds of data.

Methods

Hi Alex,

WRT the data, the grey dots are raw values, the pink dots are 10 sec averaged data as I don't have enough mechnical (air volume) or electrical damping on the sensor. The lines are post flight 'best fit' for reference.

The sensor is a PCB board



The PCB board has a differential pressor sensor and a simple op amp on it (which stretched my limited electronics skills) and gives a 1.4V to 10V output over 0-15 ish " H2O. This then feeds an anlogue input on the GRT which I've calibrated with scale and offset numbers. It is very senitive / twitchy - I'm not sure but I might be seeing the blade pluse.

This is connected to picollos....



At the moment of the three inlets (sharp edge plus) this rough hack / foam with micro balloons:


is better than a 'researched / theory' Kuchemann A-20 curve moulded and glassed in:



By about 0.5" - Go figure!

The other surprise for me was that attitude / phase seems to have a significant play in the figures, I've consistantly have 3/4 to 1 inch extra on climb out vs descent at the lower IAS - not sure if this is pure attitude or also prop RPM speed (C/S prop). The lines do seem pretty linear and cruise levels are fine. Also trying to figure out if density / temps will effect the readings along with IAS.

I'd welcome ideas....

Regards,

Carl

I did some pressure diff testing as well and found much more consistant data by using fish tank air diffuser stones as opposed to the picollo tubes. Just food for thought.

Just a reference for others, the photos Alex was refering to are here:

http://www.vansairforce.com/community/showpost.php?p=148174&postcount=8

I've sanded / smoothed further the glassed profiles (before I remove them in disgust) and back out to the hanger today for another go....

Carl

Too much chaoss below

Nice work.

Once you have a well sealed upper plenum, I think outlet flow is the area of most needed improvement and probably will yield the biggest returns.

update

Thanks Wade,

In summary I've now done some more glassing / smoothing work, set the cowl join up so the upper half slots in behind the ring and calling it good to finish. I still don't understand why the 'hacked' foam is better than the formed profile' - but that is just life. I suspect the exit area is the next focus if I want to go further down this path.

In the short term I'm going to leave the monitoring in there - I'm convinced that trying to test modification just looking at speed is not the best observation method and the variation in delta to flight profile is also key. I'm sure higher speed / lower detla could be tuned for cruise / racing and less for the climb profile - but I wanted something more rounded.

I'd still be really interested in finding the same type of data for a standard VANs baffle setup (preferablly an A model). I think electronic monitoring and recording is the way to go with this - it is too difficult to retain all the information otherwise, hard to do multiple full power climb < 1000 agl maintaining 85kts, then 90, then 100 etc.....

If people want specific details about the pressure sensors / PCBs etc let me know offline - it isn't anything particually difficult just time to research etc.

Regards,

Carl

Hummmmm

Is there a board and software that i can use with pressure transducers and my laptop....?

zkvii said:

The other surprise for me was that attitude / phase seems to have a significant play in the figures, I've consistantly have 3/4 to 1 inch extra on climb out vs descent at the lower IAS - not sure if this is pure attitude or also prop RPM speed (C/S prop). The lines do seem pretty linear and cruise levels are fine. Also trying to figure out if density / temps will effect the readings along with IAS.

Click to expand...

I really doubt that attitude is a variable, as I can't see how it could affect the pressure. The angle of attack would affect the airflow pattern, and hence the pressures, but if the weight is the same, the angle of attack in climb will be quite close to that in descent (there will be a very small difference due to any vertical force component of air from the prop changing the amount of lift needing to be developed by the wing, and hence the required AOA to develop that lift). So, I'm betting that the power produced by the prop is the variable of interest - there will be higher pressure air behind the prop when the power is higher, which seems to fit the data.

Angles and power.....

Kevin Horton said:

I really doubt that attitude is a variable, as I can't see how it could affect the pressure. The angle of attack would affect the airflow pattern, and hence the pressures, but if the weight is the same, the angle of attack in climb will be quite close to that in descent (there will be a very small difference due to any vertical force component of air from the prop changing the amount of lift needing to be developed by the wing, and hence the required AOA to develop that lift). So, I'm betting that the power produced by the prop is the variable of interest - there will be higher pressure air behind the prop when the power is higher, which seems to fit the data.

Click to expand...

Hi Kevin,

I'd love to understand more about this stuff - I'm now at a point where I'm in the 90/10 type gains - I've committed to sticking with the glassed profiles and it seems 'good enough'. I'd still love to know what the standard VANs setup gives over the speed range.

I'm happy to share the raw data if someone can get more out of it than I. The GRT has 10msec recording, which inlcude attitude, rpm, map. I also have the AoA analog output recorded and the pressure delta values.

With regard to the prop power / pressure behind the prop - do you think this is related to straight power or to prop rpm? If you can come up with a test plan - I'll happily try and fly it and provide the data.

Thanks for the comments - this is the bit of the experimental that I'm starting to really enjoy - it is learning and personal knowledge development and the international element of the help is great too....

Regards,

Carl

PCBs

Sticky1 said:

Is there a board and software that i can use with pressure transducers and my laptop....?

Click to expand...

Hi 'Sticky'

There is no 'software' on the board, it is a differential sensor plus a quad op amp and voltage regulator. I'd be ok with sharing the design (it isn't anything that special) on the understanding it is not profit / commercial use.

The key part however is what you use to record the data on. I don't use a laptop in the aircraft - do you have any 'glass' systems with logging capability. A simple analog in put is all that is needed.

Regards,

Carl

zkvii said:

With regard to the prop power / pressure behind the prop - do you think this is related to straight power or to prop rpm? If you can come up with a test plan - I'll happily try and fly it and provide the data.

Click to expand...

The prop produces thrust by accelerating air, and that higher velocity air has a higher total pressure. The amount of power is related to the change in velocity of the air going through the prop, so that change in velocity should not change much with rpm (assuming you are producing the same amount of power by selecting rpm vs MP).

If you are interested, I think you could demonstrate that by selecting two different rpm vs MP combinations that give the same power. Fly at the same IAS, and I predict that the pressures inside the cowl will be about the same. Then, do two more test points holding rpm the same, but with different MP to have different amounts of power. Do all four tests at the same IAS.

Hummmmm

zkvii said:

Hi 'Sticky'

There is no 'software' on the board, it is a differential sensor plus a quad op amp and voltage regulator. I'd be ok with sharing the design (it isn't anything that special) on the understanding it is not profit / commercial use.

The key part however is what you use to record the data on. I don't use a laptop in the aircraft - do you have any 'glass' systems with logging capability. A simple analog in put is all that is needed.

Regards,

Carl

Click to expand...

I have been refining the cowling on my seawind with a crude manometer. I wanted something I could connect to a laptop however I have an MVP-50 which will accept an analog signal. I could see all the important data this way switching from detector to detector or mabe I have several inputs...hummm I think I do.........Ideally I would like to check several areas within the cowling at the same time.....

Nope...no commercial stuff.......


Any info would be appreciated.

Thanks

Mike

Hi everyone,

I've put together a small zip file of various bits of the design board and some words to go with it. If you are keen on fabricating a PCB and what more details PM me your email address and I'll send you a link.

Regards,

Carl