AC Motor Control

[Shack]

I want to use a Picaxe to control the speed of an AC motor (fan). I know, I know but I am qualified to use the main power so that is not the issue. I already tried a 12 starter motor but it did not work ... too slow.

What I want to know is if folks think it is possible to use a relay of sorts and switch it with the PWM command to vary the speed of the AC motor? Or, a better option.

I am trying to build a device to measure the air flow thru an automotive mass air flow meter that is used on an automotive engine. They put out a signal from 0-5v based on the volume of air going thru them but apparently the data is rather secret ... so I will have to do it myself and build a chart/curve. I need a variable speed fan for this and only have an AC unit to work with.

The car computer uses the MAF to calculate volumetric efficiency and to calculate the pulse width of the fuel injector so it supplied the proper amount of fuel. Accordingly, the MAF table and accuracy is quit important. I am developing a turbocharger installation for a race engine.

Thanks

 

[Dippy]

First, have you tried a search??

Second, is it a brushed or induction AC motor?

 

[Shack]

First; I tried the search with the usual results ... lots of hits but nothing relevant. I always try that first.

Induction

 

[Tom2000]

Shack,

How much air are you trying to move? What kind of motor? How many amps?

It seems to me that you're talking a lot of air. Maybe something on the order of 750 cfm or better. That requires a husky blower, and speed controllers for big motors like that aren't cheap.

Might I suggest a mechanical alternative? Arrange your blower to fill a plenum that pressurizes your test airbox. Rig up some sort of manually-adjusted dump valve to set the pressure within the plenum. That should control flow into the airbox. (The dump valve could be something as simple as a shutter.)

It might be tricky to adjust precisely, but after things stabilize after each adjustment, you should have a steady airflow that you can measure. You might not require a precisely-determined airflow, just a steady airflow.

It will be a whole lot cheaper and easier than an electronic speed control.

Good luck!

Tom

 

[Dippy]

Tom, I thought you'd devleoped a circuit for V/F for induction motors?
Did it work or go on to the I'll-do-it-next-week pile?

 

[Tom2000]

Tom, I thought you'd devleoped a circuit for V/F for induction motors?
Did it work or go on to the I'll-do-it-next-week pile?

Dippy, I realized that I'm not a high speed, high power switching designer. I wrote a really spiffy piece of code for a 12F683 to drive the switcher, but fell flat on my face when it came to the hardware design.

I haven't given up on the idea, though. I still think, in a seat of the pants fashion, that a good hardware designer could come up with a relatively simple and inexpensive circuit to provide the switching. It's something that's needed.

When I decide to pursue it seriously, I'll put out a call for help in sci.electronics.design in hopes of finding a collaborator.

However, even if I'd completed the design, it's not what Shack needs. He needs to move orders of magnitude more air than a 1/20 hp controller could ever think of handling.

Thanks for the memories.

Tom

 

[Tom2000]

Shack,

I've been trying to come up with suitable motor/blower combinations you might use. I quickly ran through and rejected furnace blowers, vacuum cleaners, and automotive fan blowers.

I finally googled for 'flow bench' and found this homemade flow bench that uses a number of vacuum cleaner assemblies. The inventor used four for them on his cylinder head bench. That's a pretty neat idea; I only considered a single vacuum cleaner motor.

You should be able to find junked vacuum cleaners easily and cheaply. Just add as many as you need to develop your required airflow. Maybe five or six?

The rest is fairly straightforward. You can build your own manometers, and google for 'sharp edged orfice' for info on measuring airflow as a function of pressure differential across a calibrated orfice. Then some sheet metal work or carpentry to put it all together. And patience. Lots and lots of patience...

And after you've finished characterizing the MAF sensor, you have all the hardware you need for a cylinder head flow bench that you can use to tune up your cylinder heads, exhaust system, etc.

This problem is really interesting; it's the sort of problem that makes reading this forum worthwhile, even though I'm not doing much with Picaxe chips at the moment.

Good luck!

Tom

 

[gengis]

A better way might be to use the servo output to control a valve (or waste gate) in your plenum. Rotary valve operated by a RC servo that uncovers a hole in the plenum. Some felt or flocking for a good seal.

AC motor control with a relay is not practical - Most of the off-the-shelf solid state relays I've worked with include some circuitry to prevent banging the load to death with rapid on/off control. If the motor is an induction type it is likely to overheat because it will be well below its synchronous speed and pulling lots of current.

Phase control is OK with very small induction motors and someone has already posted instructions on modifying a lamp dimmer to control with a Basic Stamp. But think small motors only - 1/10 hp or less.

Vacuum cleaners use AC/DC motors (here in the states) and they can be controlled by the lamp dimmer scheme without problems. I probably saved the schematic and can look it up if you need it.

The other choice is a variable frequency drive - they can usually be controlled with a 0-5 volt external signal, but are costly. They work very well as long as you don't have a capacitor in the motor or it is a three phase motor.

 

[Shack]

I like the plenum idea with a wastegate and that would be easy ... thanks.

The fan needs to move about 275 CFM. I figure at 10 Psi of boost from the turbocharger the engine can flow about 20lb/min of air at 6000 RPM. Ignoring temperature this should be under 275 CFM. I was considering a bathroom exhaust fan since it had a 4" round outlet.

 

[Dippy]

Is a little fan going to generate enough pressure?

 

[premelec]

Dippy... no... FWIW I've tested suction on vacuum cleaner assemblies and they can reach 10psi at almost no flow so two in series might do the trick - and they [in uSA] are usually brushed universal motors which can be controlled by simple phase controllers for speed... as a first test one could - carefully - connect two shop vacuums in series and see what you've got [hopefully not a blown lid on the pressured one...]...

 

[Shack]

I have probably confused the issue with too much information. A MAF is a free flowing device like a Venturi with a calibrated orifice and a hot wire thermistor. It is setup to provide a 0-5v signal based on the volume of air flowing thru it (cooling the wire). You probably have one on your car in the air intake. There will be virtually no static pressure involved in a normally aspirated engine. A small shop vac is a great idea and more air than needed. Plus I have a couple!

 

[premelec]

OK... my bicycle hasn't got one and my long parked '69 VW bug is also deficient so I'm not aquainted with these things... I too have a bunch of shop vacs... and I hope that works out for you - note that you can use the suction side rather than the pressure side with the thingy other way 'round.

 

[Jaguarjoe]

A MAF is a free flowing device like a Venturi with a calibrated orifice and a hot wire thermistor. It is setup to provide a 0-5v signal based on the volume of air flowing thru it (cooling the wire).

A MAF measures Mass Air Flow, not volumetric air flow. A typical MAF such as on a Ford car provides an output proportional to Kg/Hr air flow.

 

[inglewoodpete]

premelec, I used to work in vacuum cleaner repair shop when I was a teenager (yes, it was a long time ago!)

If you restrict the air flow through a vacuum cleaner motor too much, it will overheat. They are quite highly strung motors, spinning at around 20,000 rpm. The motor generates a lot of heat (700w to 1500w motors are quite common) in a small space and relies on a large air flow to keep them cool.

 

[flyingnunrt]

No expert here but I reckon that...
At a flow rate of 200 cfm and a pressure head of 10 psi, (23 feet of H2O), that will be needed to replicate the mass air flow of a blown engine, to establish to voltage parameters you are after, I think you will need upwards of 10 or more vaccum cleaner motors and IF it works, then you could just turn on as many as you want to modulate the flow.
What size engine are we talking about?

 

[Shack]

NO .... this is free flow and not trying to simulate anything. Zero pressure involved.

Pushing air thru a tube with no connection on the exhaust end so to speak.

 

[gengis]

Shack

You probably should specify the pressure with some greater detail. Max flow at no restriction (zero pressure) and zero flow at some max pressure - and then blowers aren't linear.

One other consideration with squirrel cage blowers is the power requirements loaded and unloaded - a blower with a Wheel inclined to grab air will overload the motor when it is operating with no restriction - counter intuitive until you think in terms of mass air flow - with the outlet or inlet blocked it moves the air around in a circle - unrestricted it has to continuously accelerate still air to a few hundred feet a minute. You chose the type of wheel depending on how it is used . . .

I've dabbled a bit with blowers when building a pipe organ. The ordinary squirrel cage types are relatively puny with pressures in the range of fractions to inches of water pressure (U-tube manometer to measure the displacement of water since most gauges don't go down that low). I seriously doubt a vacuum cleaner motor can come close to 10 PSI with the outlet closed. And as someone pointed out - they die if they don't have air flow around the motor so a restriction isn't a good way to control pressure - but they are AC/DC and a lamp dimmer will control them.

I dug out my Grangers catalog - regenerative blowers are the only things that come close to that pressure. A single stage regen blower can actually output 110" of water at a flow of 275 SCFM (cubic feet of air at no pressure and room temperature). This particular blower is 20 horse power, $3,000, and weighs in at ~400 pounds. A 110" of water is a whooping 4 PSI. The same blower pulling a vacuum is good for only 80" of water.

For super charging they normally use a "roots" blower - which technically isn't a blower, but a positive displacement pump or compressor.

If all you are doing is checking hot wire anemometers, a wind chest or some large plumbing pressurized to whatever pressure you need with a small fan to push the air around in a circle would be the cheapest way to achieve both pressure and flow at the same time - if you need that. I'm guessing you'd also want to control the air temperature as well and that would be practical with a closed system.

 

[Tom2000]

Actually, as I see it, Shack is pretty close to being right. There's no need for any significant restriction behind the MAF sensor. In his system, the most restrictive component will likely be the test orfice.

I'd say that his air source will have to push against, at most, a couple of PSI back pressure. Probably less than that.

Tom

 

[Shack]

There is no pressure involved. It is simply a matter of pushing/pulling air thru this device and reading it’s output voltage from 0 – 4.77v.

For example this engine needs about 17.25 lb/min of air flowing thru it when the engine is running at 5500 RPM with 8 Psi of boost. All I need is the air flow to simulate that condition and then read the corresponding voltage the MAF is putting out. Neglecting temperature this is < 300CFM

A simple fan will work but I need more than 2 data points hence a way to control the fan speed to vary the volume of air. The consensus is a waste gate and I agree that is a real good idea.

I have a known device and it will be in series with the unknown. When I run my fan I can log the voltage output of both devices and then compare them to build a curve.

here is a spread sheet with a couple of known MAF's I have.

MAF Table

 

[Tom2000]

I have a known device and it will be in series with the unknown. When I run my fan I can log the voltage output of both devices and then compare them to build a curve.

That will simplify the things immensely. The most difficult part of the problem would be precisely measuring the airflow volume and temperature, then doing a bunch of ugly math to figure out the mass flow. Then repeat the measurements over and over to average them for best accuracy.

Not having to do all that hockey will eliminate the measurement instruments, including the orfice.

Yep, you can operate the system pretty close to zero back pressure.

Well done!

Tom

 

[Shack]

Yea ... I sure had some difficulty explaining what I wanted to do tho

I don't care what the actual volume/mass is.