Cooolwyp

My Feedback: (1)

I know this may sound a bit strange at first, but i was wondering if it would be possible to power each servo from its own power source? The reason for this is to determin the loading on each individual servo. I am assisting with the windtunnel testing of a 10 ft jet UAV that has 8 digital servos. We would like to try and coorilate the amp draw with the torque loading on the servo, and thus get the hinge torque and control surface moments. There are two way i have considered doing it, one being to place an amp meter on the line to each servo and then run that reading out of the wind tunnel. The other thought is that each servo could have its own power supply with a wire running out of the tunnel to the power supply. This way we can directly read the power strait off the power supply. Would this be possible? I am not sure how it has to go with the servos sharing a comon ground for the signal channel.

Thank you for your help

MinnFlyer

My Feedback: (4)

If money is not a big issue, you could use a JR Matchbox on each servo ($70 each).

The Matchbox will allow you to power each servo from its own battery (Or, for example, both aileron servos).

Here's a link:

http://www.horizonhobby.com/Products...ProdID=JRPA900

pettit

My Feedback: (23)

No problem at all. Just split the power lead (red wire) out of each servo connector and run a stout wire to a separate power supply, like a battery with a good digital meter in series with them. The negative of all these "power supplies" will be the radio common ground, the black wire.

Add another stout wire between power supply negatives and the receiver battery, at the receiver. The current will then split out to each servo.

Keep all the black wires connected between servos, receiver and battery

The big problem may be the actual reading of each meter since most digital meters sample at a rate of about one time per second. You will have to make any servo movements slowly.

The signal to each servo is also referenced to the same ground. No problem here.

Rodney

I would modify Mr. Pettit advice just a bit. What he says will work but it would be better to have the common (negative) lead of each servo directly connected between the servo's power source and the servo. You will still need the common ground wire between the receiver and servo but this is only to accomodate the signal. You will get more accurate results by having the actual load currents on a pair of seperate lines between the servo and its power source. You will have difficulty measureing the peak currents even then, you cannot use a common ammeter as that will only show the average current. You may need something like a clamp on ammeter and oscilloscope to measure peak currents.

pettit

My Feedback: (23)

Yep, now that I think about it, that would indeed be a better way.

You really don't want all the current flowing through one wire, although it does just that in any R/C system anyway.

And I agree that something better than a digital meter would be necessary.

A current probe on a decent oscilloscope will do a lot better than a digital meter.

dirtybird

My Feedback: (5)

I don't know what your budget is but Tektronix has a Hall Effect current probe that when attached to a good digital 'scope will capture almost any transient.
Not knowing your objectives I wonder if you are interested in the transient peaks. It seems to me you would be more interested in the average current. You can get that from your power supply meter.

Cooolwyp

My Feedback: (1)

We do not need to worry about any transient or peak currents, all we are looking for is once we set the position of the servo, how much power is required to keep it there, this way we know how much force is being applied to the servo, and thus can find the hinge moments for that particular control surface. Thanks for all your help everyone.

dirtybird

My Feedback: (5)

You will discover you have all kinds of problems trying to use a servo for that purpose. First the servo will heat rapidly as it stalled and has nothing to carry away the heat. The current will fall as it heats and increases its armeture resistance but the servo will hold its position because its now in holding power mode. So you will come up with several readings for the same torque depending on how long you hold it in the same position.
A better way to measure it would be get a load cell and put it between the servo and the control arm. That will give you an accurate reading of the thrust provided by the servo.

Cooolwyp

My Feedback: (1)

We would ideally like to put a load cell in, but due to cost, time, and space restrictions, it looks like it may not be posibble. We would like to get an idea of what forces the control surfaces are seeing, so it does not have to be ideal, an acuracy of 10 ounce-inches would be just fine. Also, the servo would not be stalled, we would just be reading the amperage applied to hold the servo in place.

dirtybird

My Feedback: (5)

If the servo is not moving it is in stalled condition if it is providing thrust.
Due to the high internal friction of the servo when it is in holding mode I doubt that you will get any useable information. At least be sure to use coreless motor servos

Cooolwyp

My Feedback: (1)

i was able to do some testing on the servos. I have found that there is indeed a direct linear relationship between the torque a servo is holding in place, and the load being placed on the servo. This was done in three ways, using a standard RC 4.8 volt reciever battery, using a 6 volt power supply, and using the same power supply with the servo arm at an angle. The resulting data fit a strait line with an R squared value of no less than .97. All data combined fit a trent line with an R^2 value of .9809. THe servo used was a JR DS8611a digital. If i can i will post the polts that confirm this data.

dirtybird

My Feedback: (5)

Yes you can get a plot. The trouble is when the servo heats up you will get another plot. You will need to measure temperature in order to make corrections. Unless you are working with very low torque values. Be careful with that 8611A. It will burn out easily at the high torque values. I would use the Hitec 5955. Its a more rugged servo.