PTS DIY flow sensor

[bruce]

I was sent this "teaser" pic this evening on the software side of this project.

I was told I'd be getting a working application in the next day or two for basic testing with my bench. At this time I know nothing else . . .

[bruce]

Well as of this evening I no longer just have the picture displayed above but actually have the working program on my laptop on my flowbench and it works EXCELLENT!

To give you an idea of just how well it works I used the 100 and 200 cfm pass-around-plate to see what it would do. With no tweaking anywhere the 100 plate measured 101-103 and the 200 one was 199.8 but maxing out on my sensor gain. I'll have more time Thursday to try and work on the accuracy of my configuration.

So all in all this project is progressing along and I would think it should be ready by the USA off-season "Time Build" for next year engine programs.

A BIG THANKS goes out to Tony and Rick for all the help they are giving me on this project!! I'll be the first to admit I'm just the Middleman in all of this and they are educating me as we go along!

[bruce]

Let me also add this . . . this project is not a big commercial venture it is a forum project and will stay that way. I'm going to be offering it forsale at a DIY price but, most of the support will be in the way of forum members helping to keep it cost effective and upgradeable.

The whole goal of this project was to allow anyone to get into digital gages for less money than the cost of commercial water gages and I think we are going to pull it off!

We still have a lot of little details to work out on all of this . . .

[bruce]

I took a short video of the Flowsoft program running on my flowbench this evening. Remember, this is just the preliminary program so the numbers have not been tweaked for total accuracy nor any filtering or averaging of the incoming data. What you are seeing is the raw data coming from the flowbench.

The video is about 1.4meg long





Edited By bruce on 1218159948

[Tony]

Ooohhh Ahhh pure flow bench porn Bruce.
Lovely stuff, and another great leap forward for the DAQ project.

The jumping numbers can very easily be calmed right down to any degree with some continuous running averaging in software.

This is getting really exciting.

[jsa]

Nice work Fella's.

Whats doing the pressure control in the video, Bruce ?

[bruce]

My PTS bench is setup with a Payne manual voltage control for 2 blower motor control. The numbers are jumping around as it reads the raw bench numbers. Digital gages are quite sensitive so the numbers are in constant motion, with water gages you don't see that movement but it's there.

Future upgrades to the digital manometer will be automated depression control but as of right now it's going to be simple manual control. Automated control has been talked about already between the design team. This product will be competitive in features (if not adding more) to any commercial product now on the market and keeping with the goals of my forum at a cost anyone can afford! Rick can add more to the features list of the software and he will probably be asking for input in the future as we go along.

As I stated in the earlier post there is no filtering or averaging done to the raw data. This will be programmed into the final software that goes out with the PTS digital manometer.

Our first goal was to get the data stream flowing and then go to work on number accuracy/calibration and filtering. The mere fact that we have this whole thing up to this point in a matter of a few months I think is a major accomplishment. It will proceed from this point in leaps and bounds I personally think! I'll be keeping everyone up to date on the twists and turns this project takes.

I've probably stated this many times before . . . believe it or not I'm not a big commercial entity, I'm like most of the members here in the fact that I work by myself part-time (maybe soon full-time) out of a two car garage/shop. Some of the commercial products on the market now had roots in this forum years ago and took out on their own. I know what it's like to want to do something but can't afford to do it (I'm still at that point). It's the sole reason I got into this flowbench forum in the first place. I'll continue to offer my parts at a cost effective price for everyone to purchase and "play in this addiction". I've even sent parts out to forum members who could not afford to purchase parts now at no charge so they could pursue their flowbench addiction.

[jsa]

[bruce]

Humm I'll have to listen to my bench more closely. I hear it in the video but don't hear it while running the bench. I'll have to hook the meter up to the supply voltage and see if there is a fluctuation?

I have the Surplus Center blowers in my bench right now and they "could" be the cause of the variations. They are not a high quality piece like the Ametek's are.

I'm also going to be playing with the pressure pickups on my bench and see how much effect the hole diameter has on the readings.

[Tony]

Blower surge is entirely possible with any centrifugal blower under reduced flow conditions. It can be anything from a fast flutter to a deep cyclic pressure variation. It depends mainly on the trapped air volume between blower and whatever is obstructing the flow.

[106-1194218389]

[Tony]

If it really is blower surge you will probably hear is as a sort of regular "whoofing" or "chuffing" sound in the air. Not nearly as bad as a steam locomotive, or heavy breathing, but vaguely similar in nature.

Another thing you can try is connect up headphones (or an audio amplifier and loudspeaker) to the output of the transducer amplifier circuit board. Hear what the pressure transducer hears. Thunderous waterfall like air roar is to be expected, but anything that sounds suspiciously like motor vibration or blower surge may help nail down the cause of unwanted cyclic pressure reading fluctuations.

Remember these transducers are really just sensitive high pressure microphones that pick up air pressure changes. But they can also "hear" extremely well.

[49-1183904562]

Tony;

At the point we are in now with the board and DAC laying on Bruce's bench i am not sure some of this is not just electronic noise from al those motors. The final product will need to be housed in a shielded box.

Also last night I did test some smoothing (averaging) code out on channel 1 and that really makes a huge difference. I think if we work from all sides mechanical, Electronic and Software we will end up with an accurate responsive smooth data output.

Rick

[Tony]

Rick,

I doubt it is electronic induced noise, because there is no direct electrical connection between flow bench and pressure transducer, just a plastic air line. The only really sensitive electrical nodes on the board are extremely short, and relatively low impedance, so are unlikely to be susceptible to radiated emissions from the motors.

Far more likely, is that it is direct acoustic coupling into the pressure transducers down the air line. That is not going to be a problem, unless there are some fairly large cyclic air pressure fluctuations due to vortex shedding, blower surge, motor rumble/vibration, or something similar. There is a lot of violence being done to the air in there.

Just realize that pressure transducers very similar to these can accurately track the exact pressure variations within the combustion chamber of a gasoline engine running at 10,000 rpm+ They are extremely fast, and very sensitive. They not only respond to static air pressure change, but acoustic sounds and the fastest small pressure fluctuation.

Anyhow, it will be impossible to get a completely noise free output, and funnily enough a bit of random noise will actually be an advantage for us. Your continuous averaging algorithm will interpolate nicely between the theoretical nine bit maximum resolution, and you will get a beautiful smooth output to much higher resolution than nine bits. In some analog to digital applications they actually deliberately add random noise or "dither" to achieve this same effect. But we get it for free!

That is exactly why I deliberately did not add any low pass filtering to the amplifier circuit board. Much better to keep the dithering least significant bits intact, and average in software. Any pre filtering slows down the overall response so you cannot then be quite so aggressive with the averaging program.

Don't worry about the noise as far as the project goes. All noise averages out to true static pressure, so your averaging program will get rid of whatever is there very effectively. All of us will have plenty of noise, to a greater or lesser degree. Just use the longest averaging time you feel comfortable with in your algorithm, as far as speed of response goes, and it will work fine.

[Tony]

Some further thoughts on tackling this noise problem.

It would be possible to place an acoustic absorption muffler in the air line. Maybe a glass jar full of cotton wool?

A fairly similar effect can be achieved by some electronic low pass filtering after the pressure transducer but before the analog to digital conversion.

The third method would be a continuous averaging process in software. This also works in a fairly similar way to both the above in concept.

Now the problem is, that an acoustic muffler will never get rid of all the noise, only reduce it. But it has the big disadvantage that it adds a time delay to the pressure signal reaching the transducer. The air volume of the muffler has to fill and empty before the pressure at the transducer can stabilize. So we only get rid of some of the noise, but add a big time delay.

Low pass filtering works almost the same as the acoustic muffler, we get rid of some noise, but add another big time delay.

Adding time delays in series is a disaster as far as response to pressure changes go. And the numbers on the display will still jump around.

By far the best approach to doing this is to not filter anything, go for the fastest possible responding system and just accept the random noise garbage into the software part of the system. Do all the cleaning up in software as the very final step.

That will definitely give us the fastest responding readout, with the least jumping around of the digital display.

Hope that makes sense. I have played with these types of systems for years, and this is definitely the best approach.