Pressure Sensor for Flow Bench

[garfield at pilgrimhouse.com]

On Mon, 11 May 1998 09:01:33 +1200, "Tony Bryant"
<Tony.Bryant at psc.fp.co.nz> wrote:

>As, it happens, I'm also building a DIY FlowBench. May I enquire:
>
>1) What are you using for (a) blower(s)?

Not sure this is gonna be much help to ya in Kiwilandia, but both
SuperFlow and QuadrantSci use multiples of what is refered to in the
trade as ' 5.7" vacuum motor/blowers', made by a number of suppliers,
especially Ametek, w/air-sealed bearings & double-insolated. They make a
line of two-stage blower/turbines that for example, are rated at
102cfm/90"H20, 120VAC at 9.2A [Ametek (used to be Amtek-Lamb) part #
116025-13], which sell from a big US electrical supply house like
Grainger for about $70US ea. I have looked inside SuperFlow boxes and
talked to the designer of the QuadrantSci stuff, and in both cases what
they do is of course use multiple blowers in an array, and stacked in
pairs in series electrically, which you CAN do with AC/DC motors, so you
now have a pair of said blowers drawing 9.2A/pair across 240VAC. Put say
4 pairs of these puppies together and you're up to a goodly 36A @
240VAC. Not maxing out your shop feed, by any means, but still alot of
power at 5KVA. And yes, you'd better believe it's gonna be noisy! Not to
mention the  "local warming phenomenon", heh.

>2) Can you publish your thyristor speed controller circuit, or at 
>least give us some rough details? 

Sure, got it directly outta
	http://www.teccor.com/thyristor/an1003.pdf
Look for the "double-time-constant" circuit examples. To get me up to
12A @ 240VAC (25% safety margin above the 9A I expected), I used a
Teccor Q6025J6 Triac and an HT-32 (SGS-Thomson also make the exact same
part they call a DB3) Diac. Both of these devices are available cheap
from DigiKey:
	ftp://ftp.digikey.com/Catalog/V3/PAGES/201-250/211.pdf

This is only good for driving a single PAIR of these blowers, so you
need to invent/rig either a tandem mechanical or electrical way of
controlling the pairs of blowers. If you don't wanna build these up
yerself, and plan on a manual control for the blowers, Dayton Electric
makes a similar knob-controlled circuit for fan & blower control
packaged with an integral heat sink. They charge around $100US for them!
All for a circuit that contains about $12-15 worth of parts. (Hey, all
it takes is 2 middlemen, and that's what it's gotta be marked up to).

For my purposes, controlling these pairs of thyristors/blowers via
computer, I'm gonna use some mosfet devices in place of the pots. Thas
all. (I gotta look at this 5HP 3-phase Paxton blower Bob MacKnight
recently posted on, tho; it sounds too good to be true, buy hey, greater
wonders I've seen happen via "surplus" stuff, so who knows!).
	
>I'm currently using a hotted up garden leafblower, but I'm only 
>getting 10" @ 100 CFM. (Its also very loud).

Well, that just showTaGoYa what kinda power you have to apply to move
that much air at that much vacuum/pressure. Your garden blower can't be
drawing much more than 10-12A @ 120VAC (hmm, not sure what the mains are
in Nz, but just using US figures for ducks). Multiply that by a factor
of 6-8 to get what's needed to do 20" @ 300cfm. Or compare that to the
power both SuperFlow and QuadrantSci spec for their bigger benches. It's
LOTS of AMPS @ 240VAC. Go see at:
	http://www.superflow.com/products/products-flowbench-sf300.htm
and you'll see 240VAC @ 33A or
	http://www.quadsci.com/products/engair/flwbnch/flwbnch.htm
where you will see 240VAC @ 29A, that's because they're using motor
control of the blowers, instead of 'bypass', they get some better
numbers than the SuperFlow chaps, especially at lower flow rates.

>I'm using the PWM circuit out of the back of the Motorola Thyristor 
>book, but it's kinda pulsey in this application.

Hmm, never looked in there. Not familiar.

>And a mechanical pressure sender (a modified carb 
>secondary vaccum modulator, hooked to a pot) as the feedback to the 
>motor controller.
>
>I'm using an uncalibrated flap style MAF (I will calibrate on some 
>known heads)

Alright, yer my kinda guy. It may not be "accurate" or "calibrated" but
for just comparative flow bench studies, which is the real meatNpotatoes
of the issue anyways, as long as the pressure is repeatable and the flow
delta is measurable, that's all that matters. With a small investment in
better pressure measurement, ala the stuff that
"Jack" <goflo at pacbell.net> mentioned earlier:
	http://mot2.indirect.com/senseon/mpxl5010.html
for the sales pitch and
	http://mot2.indirect.com/books/dl200/pdf/mpxl5010rev2.pdf
for the datasheets, for a 0-40"water range, where IF you can operate at
around 20"water, the linearity and accuracy of this device will be just
wunnerful for your/our porpoises. And it's "cheap", too!! (North
American Price List says $13-15 each).

The main thing to get outta all this, I s'pose, is that we're REALLY
movin a whole lotta  AIR to duplicate what a large ICE does whilst
breathing. So whether you go for a 5HP single-piece blower motor, ala
Bob McK's recent post ($350 surplus, so figure $3K new), or you add up a
bunch of smaller blowers like the big boys do (around $500 total, new),
you're still talkin about a lot of power to cover the pumping losses for
that much air flow.

Anywho, Tony, good to see there's at least one other wacko like meself
thinkin alone these lines. I bought/borrowed/read every article/paper I
could find about the older "fixed, knife-edge orifice and manometer"
style FB's ala SuperFlow, but after chatting with the designer of the
QuadSci stuff (whose name is "Oz", oddly enough), I was thoroughly sold
on this newer approach via electric control of everythang. Maybe just
the geek in me, but them knife-edge orifices ain't no easy pie to
calibrate, and have to be moved in and out mechanically, and besides, I
like to avoid the black magic whenever I can.

Just some poop and some thots, dude. 

Cheers,
Gar