Venturi Flow Element?

[1960FL]

If you are thinking of going digital you should stick around Bruce, Toni and I are working on a new project that most likely will be offered in both package and open source kit form. Just in its infancy but i am excited about it. What is your Delta P range across your element?

Rick

PS. I would recommend to all go read his build thread impressive.

[mk e]

What is your Delta P range across your element?

My slant manometer is right about a 10x so the cal data readings are inches H2Ox10

cal data Capture.JPG

Then flowing my heads, 0.05" lift is a reading of about 0.4, so 0.04" and 0.50" lift is about 30 or 3". I have a digital manometer I bought to confirm MAP readings sync TBs that I may try on the flow bench to see how it does with making the low lift readings more repeatable than my eye on the slant....the higher lift flows are very easy to read so I don't really NEED to make a change, it's more just something I'd like to play with.

I'll keep an eye out for what you guys come up with.

[1960FL]

As food for thought a 40" sensor would give you the range you need and then at 15 bit ADC you about .0012" H2O resolution/sensitivity with a mapping table to adjust for the curve of the element you could easily get that thing to produce accurate repeatable number on the cheep. I'm sure you know what to do

Rick

[mk e]

I looked at what I have, it says +/-55.4" with a resolution of 0.01. Probably better than my ability to eyeball the slant but not by much...it was about $30 iirc. I may T it in next time the bench is running and see how it does, I have no idea how much filtering/smoothing it does so it might just bounce around in a useless way.

[Brucepts]

If you are thinking of going digital you should stick around Bruce, Toni and I are working on a new project that most likely will be offered in both package and open source kit form. Just in its infancy but i am excited about it. What is your Delta P range across your element?

Rick

PS. I would recommend to all go read his build thread impressive.

Tony with a "Y" not "I" that's your Girlfriend

Glad to see you are excited about our project . . . welcome aboard

[HDgyro]

PS. I would recommend to all go read his build thread impressive.

Boy, no kidding! I dived down that rabbit hole and two hours later my jaw was still hanging open. Bravo!

I'm amazed at some of the projects you guys are pursuing in your basements and garages. Mine's nowhere near as ambitious, but I'm preparing to retire from the day job shortly and get my project in gear.

[HDgyro]

...I kind of got it working with a very fine screen cone right below the cylinder but that's long gone.

Mk e, did you also incorporate a screen or flow-straightener of any kind in the second version of the bench? Or does it rely on multiple taps into manifolds to help the manometer deal with turbulence?

I'm new at this, and I'm having a little trouble following the spreadsheet, but more of it sinks in every time I look at it. Can you explain a little about the formula you used to correlate differential to flow rates? I assume it's a repackaging of Bernouli's Law, but I'm struggling to connect the dots (excuse the pun).

I'm trying to envision a similar bench for flowing ports on a 688cc V-twin project, and I'm looking for about 27 HP per cylinder. I think that requires the bench to accommodate a max of 110-120 CFM @ 28". I plan to use two vacuum cleaner motors in a purpose-built box instead of shop vacs, to get the required depression.

Based on your experience, I may want to scale down to 3" PVC for the venturi's large pipe, with a reducer to 1.5" for the small section, to get my range of interest into the most linear portion of the calibration. The bore on this engine is 78mm, so the 3.04" ID of 3" Schedule 40 PVC would be a close match. Any thoughts? - Paul

[mk e]

The problem I had with the 1st bench was the flow would come out the head port and flow down the opposite wall of the tube, so high velocity on 1 side of the flow tube and nearly non on the other which gave bad readings...the cone shaped very fine screen mostly fixed that. I remember I tried straws but there really wasn't enough room for that to work, I still had high velocity down 1 side....just a poor design that was mostly fixed with a small pressure drop the screen caused.

The spread sheet is after I gave up on making Bernouli work.....too much noise. I just took plates that I could look up the flow for and measure them, then create an equation to connect the dots. Completely empirical.

I chose a 4" pipe because it was the closest match to the cylinder sizes I work with. That's the only reason, 3 " should be fine too, just put the neck down as close to the floor as you can so you aren't repeating my 1st bench mistakes.

[HDgyro]

Thanks for the story about the screen. Even the benches with large settling chambers use baffle plates below the test piece.

The spread sheet is after I gave up on making Bernouli work.....too much noise. I just took plates that I could look up the flow for and measure them, then create an equation to connect the dots. Completely empirical.

Wow. This is going to involve dusting off some long-dormant brain cells. I may seek you out when I get to that point!

Thanks for the help!

[HDgyro]

I've made a start at figuring out polynomial equations from x/y points, but I've used OpenOffice for years, and while it has a similar capability I think it's time to make the step up to Excel before I put any more time into it. I foresee the need to have compatibility with macros written by others in the near future.

M ke, when you built your calibration plates, did you use the standard assumptions for CD based on a very large chamber, or modify the CD assumptions at all? It looks like a calibration plate on the top of your bench would act more like an orifice plate in a tube.

Given the benefits of longer tube runs, I'm considering going in a different direction, literally - laying out a bench which uses horizontal tubing upstream and downstream of the venturi. It could all be mounted against the bottom of the bench top with everything up off the floor, and I think the initial 90º turn from a chamber below the test port could be managed.

Now, if I can just get through these last three weeks of the day job before retirement...