Dynamo for testing steam and gas model engines

[mu38&Bg#]

I think what you are saying is that the RPM or the motor will start dropping, and thus the voltage and electrical load.

Yes, that is to be expected. For A PM DC motor torque is proportional to current so as long as current increases torque will also increase. It's just not going to be linear as you add bulbs. Any load will decrease RPM. If you measure torque and RPM you can get a good idea of output power for a given pressure input. If your engine has a governor RPM should only slightly decrease until the governor can't increase engine torque anymore.

Greg

 

[jpeter]

Looks like everybody's pumped on using a generator for calculating horsepower. That being the case use the generator for the variable load by putting some type of variable resistance acrossed the output. Add an arm of known length to keep the generator from rotating. Set the arm on a small scale to measure force. Use the distance from the center of rotation to the scale times the force measured on the scale to calculate torque. I'm convenced the drag of the bearings, brushes, unknown resistance of the windings and lots of other factors are going to ball up the measurments enough to essentially make them meaningless. You could forget the generator and if the motor being tested isn't too large just set it on a rotating table with the stop arm setting on the scale.

Jim

 

[Lakc]

That #1 sullivan starter wont turn over a wet .40 engine, so I also had to go by the #2. ;D

There is a big difference in design between a motor and a generator, so the amp/load/torque values are not interchangeable directly like that. If all you want to do is measure some output and make changes to see if you have more torque or less, any type of load will do. If you want actual torque values, as opposed to just comparison values, not even the high dollar dynamometers get it that close. Calculating the parasitic losses becomes a nightmare, and you quickly find out how much temperature and atmospheric pressure variables work against you in repeatability.

 

[Lakc]

By adding load with additional light bulbs, could you just count the wattage, and assume 746 watts per horsepower?
I guess you could at least add load until there is a noticeable drop in RPM.
That would probably give at least a rough idea of what the engine can do.

You can add additional light bulbs, but you will find they all just get dimmer. You need to simultaneously measure the current flowing in the circuit, and at what voltage, to measure the true wattage.

Your actual values will be off by the frictional losses, the mechanical to electrical conversion losses, and I suspect a certain factor for the flywheel size involved as well. Thats one of the reasons a water brake dyno works so well, low friction and mass losses. (as well as load cooling)

Jeff-

I don't think there is a big difference between a motor and a generator in theory, but as you and some others have pointed out, there can be a big difference in actual application, performance, and efficiency.
As far as efficiency, it would not matter on a dyno, but it would matter on a generator, thus I would use a custom generator, not necessarily a motor to generate power.

You will have much larger mechanical/electrical conversion losses. Not a big deal if your just comparing A to B on your own dyno, but it becomes a big deal if your wondering why your ZZZ watt motor does YYY on somebodies else's dyno. Even commercial dyno operators become accustomed to that one guys dyno data will not coorelate directly to another dyno.


Edit: Didnt know there was a carlin filter on this board, replaced triple X variable with ZZZ

 

[Dan Rowe]

I would not be able to measure that much DC current,

Pat,
You simply need a suitable shunt for your multimeter. See:
http://en.wikipedia.org/wiki/Shunt_(electrical)#Use_in_current_measuring

Dan

 

[Lakc]

Here is an inexpensive hand-held laser RPM meter.
I works by applying a strip of reflective tape to the moving parts.

I have a small photocell tach for model airplane propellers, same principle and about the same price. Cant use it with flourescent light sources, however.

I would not be able to measure that much DC current, not directly anyway, but I see your point, the voltage would drop off, and the bulbs would get dimmer if the RPM dropped.

You use a small value shunt resistor, approx .1 ohm, and measure the voltage drop across that, or go to some direct magnetic reading method.

How about if you held 500 RPM, and measured it with one of these hand-held RPM meters.
Then you could screw in one light bulb at a time until the RPM meter began to drop.

Yes, although I think you will quickly develop a switch bank to just add and subract lamps. No reason you couldnt use regular incandescent light bulbs

At least then you could say this engine will run XYZ watts at 500 RPM at 100 psi, and compare it with the other engines.

Just that much information would be far more than I have now, which as I mentioned, is the crude leather glove on the flywheel.

ANY information is better then no information. As long as you have reasonable expectations, you wont be disappointed. If you cobble up something easy, you will get practical experience in what you like or dont like about the setup, and version 2 can rectify your concerns.

Speaking of steam engine model RPM, is it just me, or does it seem unusual (and dangerous) to see someone on YouTube hook a model steam engine up to air and then rev it up to 10,000 plus RPM?
I does not seem like a good demonstration of a steam engine to run it at a speed that the old time engines would never have seen.
A loaded steam engine produces a realistic sound just like the old ones, not a sewing machine buzz.
(Just my beef on over-reved steam engine models).

That one is a little beyond my current knowledge, both in youtube vids and steam engine balance, but yes, it does sound silly.

 

[mu38&Bg#]

There is no difference between a a DC motor and a DC generator other than the timing is suited for one or the other application if it is not set to zero.

I'm not going to rely on the motor characteristics for calculating power. The motor(generator) only provides a convenient way to apply a variable load. I will count RPM and use a load cell to measure torque. I'm also working on a project to generate electrical power. Matching the motor's velocity constant to the speed of the prime mover and ensuring it's capable of the current required is critical. Efficiency can be poor as long as it's taken into account.

Some fun steam links.

http://www.onthewire.co.uk/flash.htm

[ame]http://www.youtube.com/watch?v=srq_pC4v5wo[/ame]

 

[jpeter]

You've got your 120 volt bulb or your 12v automotive bulb. You can wire them in series or parallel to get a few combinations. A method of switching in bulbs has to be a real brain teaser. Treadmill motors are generally low quality 90v motors. A 120v bulb won't glow at too low a voltage. A 12 volt automotive bulb will burn out at much over 13.8 volts. And how are you going to tell with any accuracy small differences in bightness anyway. It's fun to light a bulb but I'm convenced there has to be better ways. Probably reading volts and amps to calculate watts is a good way if a comparison is your only interest. Those readings will be off too because the meter reading are not accounting for phase shift. Your meter is giving you average voltages and average amps but in lots of cases the current will be high when the voltage is maybe close to zero. Thats just the way it is with ac or pulsating dc. Put a clamp on amp meter on an ac motor and watch the small change in current as the motor get loaded. One would expect a large change in current but thats not what you see. Whats happening is the phase between the volts and amps is shifting. This shift can't be observed with ordinary test equipment.
I think I'm running on here. Help, sombody stop me.

 

[Dan Rowe]

Pat,
This is a model engine forum, the disscussion you are having with your self about KVAR and 3 phase motors really has little to do with model engines.

I think this thread should be dialed back a bit and leave industrial electrical topics to another forum. This is of course my opinion and I have no moderator hat except for self moderation.

Dan

 

[mu38&Bg#]

and the worst you can do is trip the breaker feeding the motor.

Unfortunately, this is not the worst thing you can do. The big issue with feeding power back to the grid illegally is not disconnecting from the grid when the grid is down.

Let's just end the grid tie discussion here. No need to delete a whole topic. If you want to edit the grid tie stuff out ask a moderator for help.

Greg

 

[Dan Rowe]

Pat,
I never said this is not the proper forum for you.

Industrial power is a subject best left to the professional type forums with proper reference to legal code and standard practice.

I think you have a lot to offer the forum and I sincerely hope you stay on board.

Dan

 

[SAM in LA]

Dan-

I have deleted all of my posts with regards to this topic.
Please let me know if you find any of my other posts objectionable, and I will remove those also.

The very least of my intents on this formum is to post anything that would upset anyone else in any way, or be in any way inappropriate for this forum.

I suppose you are correct, this is not the appropriate forum for me.

Pat,

I really haven't been following this thread closely and I do not know why you would think that this forum isn't a good place for you. But I will, without hesitation, say that you should stay with us. You have contributed so much and have so much more to offer that it would be terrible loss to all of us if you left.

Highest regards,

SAM

 

[jpeter]

Sorry. I just wanted to add to the nonsense. I guess I got carried away.

 

[Dan Rowe]

I'm not going to rely on the motor characteristics for calculating power. The motor(generator) only provides a convenient way to apply a variable load. I will count RPM and use a load cell to measure torque.

Greg,
I checked load cell on the web and found the Futek site but I have never encountered these devices. Can you say a bit more about how to use a load cell to measure torque?
http://www.futek.com/product.aspx?t=load

Dan

 

[mu38&Bg#]

http://www.futek.com/product.aspx?t=instrument&cat=bt&m=ibt500

You need one of these. Or some other way to take the small signal and amplify and scale it to a meaningful reading. I'll be using a data acquisition card. Digital scales use loadcells. If you have one, take the covers off and look inside.

http://en.wikipedia.org/wiki/Force_gauge

Once you have the load cell calibrated you take into account the moment arm of your dyno and can calculate torque. Think of it as using a digital scale on the torque arm of a prony brake.

Greg

 

[jpeter]

I once used a prony break to measure the output of a 60 size model airplane engine. The moment arm was the engine mount. It worked ok but the scale jumped around a bunch. Had to conclude an average reading. I was using a spring scale. I wonder how a digital scale would have worked. Might be hard to stabilize it enough to get a reading. I think the spring scale would be best.

 

[Lakc]

I once used a prony break to measure the output of a 60 size model airplane engine. The moment arm was the engine mount. It worked ok but the scale jumped around a bunch. Had to conclude an average reading. I was using a spring scale. I wonder how a digital scale would have worked. Might be hard to stabilize it enough to get a reading. I think the spring scale would be best.

A digital scale isnt that much different from a spring scale, except the dampening is done by software instead of some mechanical means.

The origional issue of the SIC water brake dyno used a diaphram and a pressure guage, istr reading somewhere that they eventually went to a load cell device.

My origional dyno was for model airplane motors, I simply used the same prop on all my tests and used printed values for torque/rpm. My dyno did get me to program my first microcontroller, as I needed a way to count the output pulses from a automobile airflow meter. If you learn micrcontroller design and programming, you can use one to totally automate and log the entire measurement process.

 

[Dan Rowe]

Now I get it a load cell is simply a packaged strain gauge.

Now I have heard of strain gauges and I have even seen one super glued to the shaft of one of the Sealand Econ Class of ship. A really big hose clamp was used to attach the battery powered device with a radio transmitter. The device could also detect the shaft RPM, so the data could simply be converted to shaft horsepower.

Links for strain gauges and circuits:
http://www.rdpe.com/ex/hiw-sglc.htm
http://www.omega.com/literature/transactions/volume3/strain2.html

Dan

 

[Lakc]

Yep, if that peaked your interest, here is a well done homebuilt one with a lot of good info.

http://www.nakka-rocketry.net/strainlc.html

I am not to be held responsible if you begin to lust after making homemade rocket engines. ;D