PTS DIY flow sensor

[bruce]

Over the past couple of months since the release of my plans the "flow computer" process has been a sticky point with me as it has for other members of this forum. So an undertaking was begun to design and build a DIY (?) style flow computer (I use the term flow computer but it's really not a flow computer per-see) Since I had no commercial products in my shop to "look at" nor would I have done that, there is no fear it is a copy of something already on the market.

With the help of various forum members behind the scenes I can now say we are getting closer to a final circuit. things have really progressed in the past month after receiving a schematic from forum member Bill Shaw, he sent me a schematic that we would be able to use as a base for our project. With the help of Tony, I decided I might be able to actually pull this off with my LIMITED electronic experience.

I can say that I now have a single sensor circuit working in my shop. The plan calls for this sensor board to interface with a DI-194RS card sold online for around $25.00, this will then feed the data coming from the card to Excel for data gathering and manipulation.

The final plan is for the sensor board to have 3 pressure sensors, 1 for static, 1 for differential and one for a velocity probe or another pressure connection. We already have a working schematic which has been extensively modified from the original copy I received from Bill.

We are working on a bill-of-materials (everything can be purchased from Digikey) as well as pre-made PCB's for the components which will more than likely be offered here on the forum for a reasonable cost. The circuit could also be breadboarded if you are not handy with a soldering iron. (I might offer preassembled/checked boards). If there is enough interest I would also entertain buying components in bulk at a cheaper price and offer them here as a package deal for final assembly?

This sensor board will use Larry C's FPexcel product or something similar or you can build your own spreadsheet. My thoughts are for this project to be supported here on the forum by it's members for it's members and not a totally commercial product! That being said there might/will be some end user tweaking possible so keep that in mind as it is DIY.

This project is separate from the one Daniel (MrHijet) is also working on and should in no way effect his efforts which I highly support. We will possibly end up with a few different designs of digital gages from which all members will benefit from.

A "rough" idea on a cost for this project is going to be somewhere around $150-200 range depending on how one approaches it.

At this point in time I really don't know my personal direction from a forum stand-point on how this is all going to playout. I just wanted to let everyone know something is being worked on.

[Tony]

[bruce]

The PCB prototype boards Tony developed for this project showed up today!

[bruce]

Here is some more "eye candy" to look at for anyone who is following this thread.

Disregard the wires running all over the place the board is still being prototyped and I'm running some parts off the breadboard for testing purposes.

[Tony]

[bruce]

Here is some more "eye candy", a shot of the digital manometer hooked to the DI-194 on my bench:

[bruce]

A shot of the screen with the generic DATAQ software output showing channel one (Static) and channel two (differential) it's tough to see the screen:

[MBateman]

If you need a serial port for a computer that doesn't have one they do make usb to serial adaptors. We went through several on our dyno computers before finding one that works. As far as I know the only one that really works is the Keyspan 19-hs.

Michael

[Tony]

Bruce's system uses a simple DIY circuit board, plus a very low cost commercial data acquisition unit to get raw digital pressure data into a computer.
The advantages of this approach are that, assuming you already have a suitable computer, it is absolutely the simplest and lowest cost approach possible. Once the numbers are entered into a spread sheet, some suitable number crunching can then be applied to do some really wonderful things.

Danel's system is similar, but it will have a rather neat inbuilt alpha numeric display panel, and an internal microprocessor. It will therefore cost more to build and be a lot more complex. But it will have the advantage that it can possibly be run on batteries, and be used portable, alone, WITHOUT an external computer for simple direct pressure monitoring. Great for taking multi pressure readings in a moving vehicle for example. This system will also input numbers via a data cable into an external computer (and spread sheet), so it is really the deluxe version.
That is as I understand the plan at this stage. Being still in early development, Daniel may have further ideas.

There is definitely a place for both approaches, and one may be more suitable than the other depending on planned usage.

Bruce's system is a three channel differential pressure acquisition front end for an existing computer. It follows the same philosophy as the PTS flow bench, in that it is a 100% DIY project for the budget minded that gives away nothing in performance.

Daniel's system is a stand alone portable multi channel digital manometer with inbuilt readout, that can also be plugged into a computer if required.

[Tony]

Doug,

The PTS system uses a circuit board which contains three differential pressure transducers, and three high gain amplifiers. That is all it does, there is no microprocessor, or anything really complicated.
Simple, low cost, and very easy for anyone to copy.

It was assumed that this will be used mostly with a flow bench, where there will always be mains power available, so it was decided power from a dc plug pack would be the most convenient. There is no reason why this circuit board could not be powered from a pair of small nine volt PP9 batteries (connected in series) in a race car.

The data acquisition module is powered off the RS232 data line directly, so your laptop would supply the power for that.

There are three differential pressure transducers, each transducer having a "high pressure port" and a "low pressure port". The transducer measures the difference in pressure between the two.

Vacuum can be measured by connecting to the low pressure input port, and leaving the high pressure port open to atmosphere.
Positive pressure can be measured by connecting to the high pressure port, and leaving the low pressure port open to atmosphere.

But the really exciting thing, is that you will be able to measure small pressure drops directly. For example, air pressure drops across radiator cores, air cleaners, pipework, intercoolers, mufflers, or throttle bodies.

The pressure transducers recommended for use with a flow bench will most likely be the 1.5 psi units (42" water), but similar transducers are also available in more sensitive ranges, or for higher pressures.

The three pressure transducers and the three associated amplifiers are all independent. The three can be given different pressure ranges if required. The gain of the amplifiers can be changed by changing one resistor value so that a pressure transducer can be scaled to have any convenient full scale sensitivity (within it's pressure reading capacity).

For example, a 40" transducer can have the amplifier gain set so that the output reaches the full +10volts at 40" or 25" or 13" or any other pressure, by suitably changing the amplifier gain.

[bruce]

Tony, thanks for explaining the circuit!

Doug, you can setup a Clippard 5 port valve to swap the sensor from intake to exhaust mode. This also will work with the static reading. 3 sensors were decided upon to read static, differential and one for a velocity probe (since I sell them) the third sensor could be used for fuel signals or something else but it only has a range of ~40" H2O.

As Tony said other Motorola MPX20XX series sensors could be used in place of any of the sensors we used to work at higher ranges. Since the DI-194RS only has 4 inputs we did not add any additional sensors. If you go with a higher end DATAQ board with more inputs you could use 2 boards for more sensor inputs. Tony was able to design the PCB keeping it within small footprint so the cost to get a PCB manufactured was well within the price range of a DIY project.

The goal was to get a simple and "cheap" digital setup to compliment my PTS plans which I feel I am 50 to 75% of the way to achieving, once we have the DAQ to spreadsheet/database part working it will be 100%. As we all learn more about digital setups we can expand on what we are doing and I'm sure other thoughts and ideas will come along. There is already thoughts on a flow depression control, but it is just thoughts at this time.

Right now the commercial units are charging more to add on a third sensor to their 2 sensor product as this whole project has cost me to date! Since we will be documenting this project somehow on the forum the end user will be able to control some of their design, might be a "bag of worms" but hey at one time a flowbench was "Black Art" . . . and look where we are with that now!

[115-1172523331]

Tony, Thanks again! I thought I understood the PTS system so it confused me when you said that "Danel's system is similar, ... But it will have the advantage that it can possibly be run on batteries...". I didn't see why the PTS couldn't also run on batteries and you have now confirmed that it can.

Remember that I have a variable inclined manometer intended to supply more sensitivity when I ask the next question. Since changing the value of a single resistor changes the gain, could a variable resistor be put into the circuit at that point so a "variable sensitivity" transducer would result? I would have to be able to determine the resistance value of the setting being used (which could be a spreadsheet input as height is now) and have to have some "faith" that there is no drift, but ...? And to think I never liked electrical engineering when I was in college!

Since Bruce reminded me of the Clippard valves (Thanks, Bruce), there is no need for reversing intake for exhaust manually, or the 4th sensor.

There are 4 channels on the DATAQ so is the 4th used now, or is it available for some other crazy idea I come up with? I'm really looking forward to seeing this in action. This is too much fun!!! -- Doug

[Tony]

Doug, when I said the systems are basically similar, I meant only with respect that both measure pressure, and both can send serial data to a computer. At the most basic level, that is what they both do.

But Daniels will do far more than that !

Yes, you could use a variable resistor to adjust gain, in fact it is a fun way to experiment. The problem being that this resistor will be pretty vulnerable to small variations and upsets. For best long term and temperature stability, it absolutely must be a fixed resistor of a very stable type.

The idea is to roughly scale the pressure transducer output voltage so that it uses almost all of the available DAQ input range for best possible resolution. Fine final calibration is then done in software.

This resistor will be soldered between a couple of vertically projecting pins on the circuit board, so it can be very readily changed many times without any risk of damaging the tracks or pads on the board.

Batteries are a mixed blessing. They invariably go flat at the most inconvenient times. If mains power is available, a plug pack will always be preferable for something like a fixed pressure monitoring system built into a flow bench.

But if portability is your thing, it could be built into a plastic box with a battery and on/off switch. Just remember to turn it off.

[115-1172523331]

Hi Tony, I understood that Daniel's system will do much more than the PTS system (at a higher cost and more complexity). The extra parts (LCD, microprocessor, etc) have to consume more power, that's why the "advantage" comment you made about batteries confused me. I thought I might be missing something on the PTS design requirements. I was the Manager of Advanced Aircraft Design at Rockwell (bought by Boeing) when I retired and spent a good deal of my time looking for "alternate applications" for new (and existing) technologies. Like a 3 year old, my favorite response to "Here is what we HAVE to do ..." was "WHY?" It made for some "interesting" discussions but sometimes also some very interesting results! They say a little bit of knowledge makes one dangerous. I was VERY dangerous! Thanks for keeping me that way! -- Doug

My primary problem with systems is forgetting to turn them ON rather than OFF. I was told that "everybody has done it" when I didn't turn the fuel petcock ON before making my first pass at the first Muroc Reunion. Everybody may have done it, but I did it in front of 10,000 spectators! I now have a check list on top of my speedometer.

[Tony]

In developing this circuit, the possibility of battery operation never occurred to me, hence it is rather power hungry.
But if the LED is removed, a pair of heavy duty PP9 batteries connected in series (18v) should last several hours, certainly enough to do some road or rack testing.

My original thoughts were that if there is power to run a flow bench, then there will be power available for a dc plug pack. So there will be plenty of surplus power available to run a nice bright LED.

While battery operation is possible, it really draws far too much power to run it from batteries all the time. I never anticipated other applications beyond using this manometer in a mains powered home flow bench.