CIM speed in reverse?
 

My team just finished building and mounting our first attempt at combining the Boshe Drill motors and the CIM aka Chiaphua aka Atwood motors. We matched the free load speeds of both using gears and some chains, and it tested fine while we had the robot propped up on a table, but when we placed it on the floor, one side was completely overpowering the other side and it was impossible to maneuver the robot.

The way our gearbox is set up is that both the Bosh Drill motors turn forward when the robot went forward, but due to lack of space and eagerness to see the robot move under CIM and drill power we had 1 CIM go forward and 1 CIM go back wards, hoping that the RPM difference was not too great in forward or reverse. Can this be the problem that we are facing? That the forward CIM is going so much faster then the back wards one that it is dominating the other side?

Thanks!

Re: CIM speed in reverse?
 

http://www.chiefdelphi.com/forums/sh...ad.php?t=23845
http://www.chiefdelphi.com/forums/showthread.php?t=2140

From what I can tell in your post, the CIM shouldn't be causing your problem. Do you have any pictures or CADs of your setup? This might help us try to figure out your problem.

-Bill

<edit>
Mike brings up a good point about the possibility of breakers tripping in his post below.

There are any number of possible reasons for this symptom.
</edit>

Re: CIM speed in reverse?
 

Paranoid,

Yes, the CIM is different in forward and reverse. However, due to the magnitude of error you are experiencing, I'm guessing there is something more fundamental involved.

Try and disconnect the power between the drill speed controllers and the drill motors (run just the CIMs). Repeat with just the drills. I'm guessing that this test will provide more illumination.

Also, pay attention to the circuit breakers, one may be tripping on you.

Re: CIM speed in reverse?
 

The CIMs are, in theory, pretty symmetrical (no huge angle difference as in the drills), so that should not be your major issue. You even made both drills turn the same way, so your motors are not likely to be the problem. Look for gears misalignment and, above of all, different chain tensioning from side to side.
Other factors might be shaft bending under load (robot OK on the table, bad on the floor) and, less likely, uneven weight distribution between the left and right sides.
Do you identify any of these in your design?

Re: CIM speed in reverse?
 

i don't believe we even need the cim's but they still want to use them, and sry, but there will be no pictures of our setup, sry for the dispointment, but thats top secret :rolleyes:

hmmm, it could be the chains, but i don't believe that's the main problem being it worked fine until we used the cim's, but tahnks for oyour help guys

Re: CIM speed in reverse?
 

Another thing to check, make sure that the drill motors are both shifted into the correct gear tightly. I know this may sound dumb, but I actually had a problem with this and it took me a long time and a few posts on here to figure it out. I'm probably the only one dumb enough to do that but I just wanted to throw the idea out there. Good luck with getting it straightened out.
-Aaron

Re: CIM speed in reverse?
 

Here are my teams experimentaly determined numbers for RPM.

http://chiefdelphi.com/forums/showpo...0&postcount=13

It sounds as though we did basically the same thing you are doing last year, and it worked wonderfully for us. I don't think its possible that your issue is being caused be the CIM motor differences themselves. More likely are uneven chain tensions, bent axles, or a too-tight center distance on one gear set. Before you put the motors in, did one side turn more easily than the other? Is one side now noticably more difficult to backdrive by hand than the other side, when the system is not even turned on? (if you didn't remove the internal pins in the drills, maybe don't try that)

I can say that, as a driver, adding the CIM motors to the drill motors made a huge improvement to both the pushing power and manuverability of our six wheel drive robot. I am a big fan of quad motors drives because they are within the reach of almost any team and are still competetive with more complicated shifting drive trains. I would not give up on this experiment, especially given than its the off season.

Re: CIM speed in reverse?
 

Check that each individual motor is functioning. We had similar problems and it turned out one of the motors wasn't running, I think it ended up being a wireing problem.

Wetzel

Re: CIM speed in reverse?
 

The free load speed? Err....isn't that a contradiction. If you meant the speed they spin up when there's no load (I.E. Free Speed) than that's why one sides overpowering the other, you need to match their speeds while under load, if you meant you matched them under load, then..I don't know.

Hope that sheds some light on things.

Re: CIM speed in reverse?
 

As far as I know, the CIM motor has NO winding-bias, and spins with the SAME characteristics in forward and reverse.

The variance between drive sides probably has something to do with your gearbox. Possibly one has a parasitic load somewhere you don't know about, which accounts for it running slower.

John

Re: CIM speed in reverse?
 

No you don't... people match the free speeds all the time. As long as both sides are matched to the free speed, they'll be moving at the same speed... whether that's the optimal place to match the speeds or not is a different question.

Re: CIM speed in reverse?
 

Could you describe your control system? While I agree that a mechanical problem is the most likely cause, I wouldn't discount the possibility of a problem with the joystick(s) or even one of the victors.

Have you calibrated the victors? Are the victors all running? Are you sure that your programming is legit? It's possible that one motor isn't being powered at all. It's actually a very easy thing to have happen, and you probably wouldn't notice with no load on the system. All it takes is one of the breakers to come loose or a PWM cable to not be fully seated or any number of other things.

Just one more thing to check on. As long as all the victors are lighting up green when you have the stick down, then it's probably not a control system problem.

-Andy A.

Re: CIM speed in reverse?
 

Wow, lots of disinformation in this thread. Let me give you some facts.

1. The CIM motor windings are symmetric, so there is no inherent speed difference between forward and reverse. However, there are small differences from motor to motor due to manufacturing tolerances but not enough to notice. We have tested these motors extensively and they have the same behavior in forward and reverse.

2. Matching motors at free speed is perfectly fine. Matching motors at any other speed is fine, too. Please note that if you match at a speed lower than your top speed, then any speed over that matching speed one of the motors will be doing negative work. What this means is that one of the motors will be acting like a generator and not like a motor. What this also means is that one of the motors will be doing all the work when running at a speed above the matching speed.


Now for some of my opinions on your drive problems:

1. The fact that your conditions under load are different than sitting on the table points to either your gear box or drive train. Bent shafts, bad bearings, gear or chain misalignment could all be the culprits.

2. It could also be a bad connection, but usually you can find that problem by just running on the table. To find out if it is a bad motor, systematically disconnect one motor at a time and see if the side you disconnected the one motor from still drives. This happened to us a few times and this method found the problem each time.

-Paul

Re: CIM speed in reverse?
 

Sounds like there is an answer somewhere in here.

Our normal debugging goes like this:
1). Is it software? Always blame the software guys first....
2). Check that each motor is working by itself. (a wiring problem)
3). Make sure that the motor pairs are driving in the same direction. (a wiring polarity problem)
4). While propped up on a stand see if you can see if one side is running faster than the other. (a gearbox or drivetrain problem).

After that, you should have a good idea of where the problem lies.
We have been using the drill motors and CIMs combined for several years now and have not seen a problem like you describe.

Re: CIM speed in reverse?
 

First, Thanks for all your help guys. Honestly if you guys did not respond, we probably would have tried to mount the motors so they are both facing forward and thus wasting time because it wasn't the actual problem.

It was, in fact, a hardware problem. We went though the various ideas that you guys posted and it turns out the victor controlling the right motor was not functioning, so we swapped it out and everything worked like a charm.

:) Thanks Again :)

Re: CIM speed in reverse?
 

If it worked right on the table but not on the floor I think you still have a problem. The dead Victor may be a symptom and not the disease. Anytime a drive works in the air but not on the floor, start looking for misaligned parts in the drive. Bearings that are not parallel ( on two ends of a drive shaft) is often the problem. Those that use chains often put too much tension on the chains. Remember there are very few of the kit motors that can stand any side load and therefore need to be coupled to shafts and bearings that can take the load. i.e. Don't put a chain sprocket on the shaft of an FP, drill or Chalupa and expect it to last.

Re: CIM speed in reverse?
 

Al,

In general, you are correct. However, for this particluar case I think the Victor was the problem. They are using the drills and CIM motors together matched at free speed. If only one victor was bad, then one of the paired motors would still work. While in the air (aka almost at free speed), the side with one motor would still look like it was running with two due to the matching. But once under load, the problem becomes clear. This exact method is how we have found bad Victors (or disconnected wires) in the past.

-Paul

Re: CIM speed in reverse?
 

Paul,
Wouldn't there have been a noticeable difference in speed if one of the motors were not driving, even in no load? I guess the effect would have to depend on which motor was nonfunctional vs. it's ratio to the final output.

Re: CIM speed in reverse?
 

Al,

You would think that the one side would move slower due to the fact that one motor had to actually drive the dead one. The fact is, even though it does move slower it is very hard to tell by eye if it is slower. If I understand where you are going with this, you are right that this problem could have been detected on the table.

This next part is to everyone else (Al's team already does something similar to this and they probably taught us this at one time),

A procedure we do every year is to check the current draw of each motor when the drive base is up in the air. It is rather easy to do. Simply put an ammeter (or multimeter set to current) in series with the motor and measure the current draw of each motor. The CIMs on each side should draw the same current and the Drill motors from each side should draw the same current. You would have found the Victor problem from this test. We do this test to make sure our gearbox is the most efficient it can be and to make sure we do not have a bad motor, bad connection, bad speed controller (Victor).

If the above procedure was done while the robot was in the air, then you would have noticed the problem.

-Paul

Re: CIM speed in reverse?
 

Using a multimeter was how we actually found the problem. The victor was tricky :rolleyes: ... light were lighting up right but when we tested the current going to the motor it never changed when we moved the joystick...

i of course blamed the code for it couldn't be my wiring :D , but thanks guys, it always does turn out to be something silly

Re: CIM speed in reverse?
 

Sorry about the delay in reply. This is really an unusual failure for a Victor. Can you detail how you measured the current? Were you tripping breakers? Do you still have the controller or did you return it to IFI? I would like the chance to check it out.

Re: CIM speed in reverse?
 

Al is right!
I've never heard of a transistor jumping to saturation before. have you calibrated that victor lately? Al, if you get any info, tell me please!!!

PS: Read my signature. Amazing as it sounds, many electrical failures were because of it, and I have seen some weird things when I fix stuff.

Sparks

Re: CIM speed in reverse?
 

1.) We measured the current by propping the bot up on boxes and running it forward, then we hooked up a micrometer to the 2 screws that went to the motor and it was reading 0 volts. We checked the code and everything was in place.

2.) I really doubt we were tripping the breakers, there was no resistance since it was propped on boxes. Plus when we installed the new Victor it worked, so that was the problem.

3.) Yes we still have the Victor, actually we put it back in the working victor box for unknown reasons. If you would like to check it out, PM me details to send it to you.

Re: CIM speed in reverse?
 

The Bobcats have encountered something similar. We had a few issues where a Victor gave us no output. We would see a flashing LED (as if the controller was not powered).

We replaced the Victor and everything was OK. However, we checked the "bad" Victor and it would check out OK... Further checks and we found that the PWM cable was not seated fully down in some of the Victors.

This was NOT a bad cable problem. We would install a 1' PWM extension cable at the Victor end and the problem would go away.

Does this sound familiar to your problem?

Re: CIM speed in reverse?
 

Not really, although I am not 100% sure, I think that the LED lights were lighting correctly according to the joystick movement.

Re: CIM speed in reverse?
 

Colt,
A multimeter might not measure any output voltage on the Victors until they are at full throttle. (Indicated by a steady RED or Green LED) Depending on the meter you use it may not be able to measure the pulsing AC waveform sent to the motor. If you were at full throttle then a DC meter should indicate approx. 12 volts. An oscilloscope would indicate whether pulsed output was available but be careful to not ground the output of the Victor when using a scope. Connect the ground lead to the "-" screw on the input to the Victor and then probe either of the output terminals for a display. Only one will show output this way for forward and the opposite one for reverse.

Re: CIM speed in reverse?
 

Ah, yes, I forgot to mention that we pushed the joystick forward full throttle with the LEDs solid colors when we were testing the volts, which read 0v.

Re: CIM speed in reverse?
 

Al,
I think the victors put out DC.
They take in PWM and DC, and the motors are DC, so I don't think they'd have AC coming to the motor. I think it's simply a steady variable voltage going to the motor. Pulsing a motor will shorten its life considerably.

Sparks

Re: CIM speed in reverse?
 

The speed controllers do not have the room for a smoothing inductor capable of such amperage. The inductance of the motor provides for a rather smooth DC current but the DC voltage is chopped at a fairly high frequency.

Almost all variable speed motors (DC or AC) use chopped (PWM) waveforms.

Re: CIM speed in reverse?
 

Sparks,
The controllers are capable of pulsing battery voltage to the motor at full level with a PWM type signal. As the length of the pulse gets longer the average current in the motor goes up. At full throttle (when the LED is "on" steady) is the only time that the controller puts out a pure DC signal. If the controllers were to put out a varying DC voltage they would have to dissipate the difference in voltage between what the battery puts out and what you want to send to the motor. At low throttle near stall, that could be considerable heat to get rid of. One of the nice effects of this operation is that the motors are pulsed at full voltage and therefore easily overcome starting frictions and torques. I can go into a more in depth desciption if you would like. They are really pretty cool little animals.

Re: CIM speed in reverse?
 

I might as well add in something about variable speeds through variable voltage...some folks in R/C car racing have dealt with a Manual Speed Controller (MSC). Essentially, it's a servo with an arm that feeds power to the motor and a set of resistors, doing exactly what Al described (giving X amount of voltage to the motor, then the rest is dissipated through resistors that create heat.

I've personally never used one (the ones I dealt with seem to have a tendency to release the magic smoke, according to my sources), but the fact that you're feeding a fair amount of your battery power through resistors simply to make that excess power into heat is grossly inefficient. Dangerous, in some cases. (That, and having taken one out of a truck I bought off eBay, they're obscenely bulky. Seriously--if weight is the enemy, these things have pictures in all the post offices.)

Hence roughly 99.999% of hobby-level R/C cars (and all of the racing versions) use Electronic Speed Controllers.

Just figured I'd give a practical example--the tip jar is at the door. :D

Re: CIM speed in reverse?
 

Al,
Thanks for the offer, but I know how they work. I got bored one day and looked them up.
Furthermore: You are correct in saying that they don't put out DC, persay, but I wouldn't say they put out AC either. AC as most think of it is a voltage travelling from a plus voltage to a minus voltage, and back, all at a frequency. If you hook that up to a DC motor, sparks will fly. Guaranteed. In conclusion, AC motors are hard to vary speed to (they use different windings for different speeds.) Victors do put out PWM, or, as I like to think of it, modulated DC, and I am sorry for any confusion I may have caused. Also, barring the fact that there isn't much in the way of alternatives, I think it's a really stupid way to vary motor speeds. Like I said before, it can shorten motor life considerably.

Sparks

Re: CIM speed in reverse?
 

Sparks,

When you try to control voltage in the analog world, you end up dissipating a lot of power.

As an example, let's assume that you are delivering 3V at 20A to a load from a 12V source. The power dissipated in the load is 60W but the power delivered by the source in 240W. 180W of power is dissipated in the analog speed controller in the form of heat.

A perfect switch is either open (full voltage, no current) or closed (no voltage, full current) which makes the power dissipated across it 0W. Yet with a 25% duty cycle, it can deliver the 60W to the load with only 60W delivered by the source. Of course, a power semiconductor is not quite a perfect switch, but it is close.

Without semiconductor power electronics, one of our 130 pound robots (with the same power delivered to the motors) would discharge the Exide battery in less than a minute and the 6 AWG wires would have to be 1/0 or 2/0 wire. Also, the robot would weigh over 300 pounds (the controls and wire would weigh more than the mechanical systems).

In short, we could not do what we do in FIRST…

In reference to AC motors, a variable speed AC induction motor requires both amplitude and frequency modulation. Without power electronics, making a variable speed AC motor required a motor/generator pair with a variable speed transmission coupling… Very inefficient.

Now, let’s talk about the effect of a PWM wave on a motor. Yes, it has an effect. However, that effect is mitigated by the fact that the PWM frequencies are high and getting higher with every new generation of power semiconductor devices. The impact on a motor is a function of the harmonic frequency of the switching due to the inductance of the motor. This falls off at 20 db per decade. Bottom line, there is some effect but it is minimal.

Let’s look at it from the perspective of music. When the CD player came along, many of us unwittingly destroyed the speakers of our stereos because the dynamic response of the digital music was beyond the ability of the speakers. Would you say that digital music was stupid and we should still be playing vinyl record albums? I think not… The industry began making speakers which could withstand the new requirements.

Without power electronics, we would not have portable personal computers, alternative power co-generation and flat panel TV sets (not to mention space stations, shuttles and communications satellites). It is a basic enabling technology second only to the transistor in its impact on society.

Stupid? Not in MHO… If you can invent a better way of doing it, patent it. There should be a Nobel Prize in it and you will be wealthier than Dean within a year!

Regards,

Re: CIM speed in reverse?
 

Sparks,
In our real world, anything that is termed DC is considered a steady state or non varying electric potential. Any voltage that varies is called AC, that does not imply it is a sine wave, just that the potential alternates.
Please read Mike's post above as he very nicely explains analog and digital controls systems. As to hooking up AC to a DC motor creating sparks, most wired drill motors are termed AC/DC motors. By their very nature, the current flowing through the motor windings can be either DC or AC. In our case, the drill motors and other motors are all DC motors. We are pulsing that DC on and off at a predefined rate and since that rate is fairly fast, the laws that hold for AC circuits holds for our circuit analysis as well as DC laws. For instance, a scope looking at the output of the controller will display the signal either in DC or AC coupling. An AC voltmeter will measure a voltage where it will not measure a voltage when connected to the battery.

Re: CIM speed in reverse?
 

I forgot to mention this...Using a speed controller does not shorten motor life. The sparks you see inside a brush type DC motor are caused by the collapsing magnetic field in the armature. As the brush moves off the commutator segment and the coil goes open, the magnetic field set up by the current flow through the coil immediately starts to collapse. This "AC" (since it is changing/collapsing) flow induces a very high voltage and that is what you see arc across the brush and commutator. Higher loads produce higher currents which produce higher voltages when the brush circuit opens. Just look inside any drill while you are drilling holes. You will see the arcing inside.
This is the same method spark voltage is generated with an ignition coil, a high voltage pulse is generated when the distributor points or electronic switch open the current supplied to the ignition coil. Oddly enough, this arc produces electromagnetic energy and was used in the early days of radio for transmission. Anyone who has heard the story of the Titanic knows that a radio was in use and that transmitter was a spark gap device.

Re: CIM speed in reverse?
 

Al,
I am not suggesting that the sparking has anything to do with the motor life. If one pulses the power, you shorten motor life, for the same reason that turning a light on and off repeatedly shortens life: Going from no power to full power in a very short amount of time is damaging. For example: On an RC helicopter, a standard motor life is about 50-70 hours, using typical speed control. On a new helicopter that uses a gyroscopic swashplate (it achieves this by pulsing the motor at specific intervals), the motor only lasts about 10 hours. I think this is an example of how motors are damaged by pulses. That is why I like analog voltages.

Sparks

Re: CIM speed in reverse?
 

Sparks,

Why don't you explain how a "typical speed control" works...

Re: CIM speed in reverse?
 

Sparks,
I need to respond to this so others will not be misled. Pulsing power to a motor does not shorten motor life. If you were to analyze DC motors, you would see that the brush and commutator already are pulsing current to the motor as normal operation.
A filament in a light bulb is a far different device. The metal in the filament has more than a thousand degrees thermal difference between on and off. As this rapid temperature change occurs the tungsten rapidly contracts and expands and it is this flexing that eventually breaks through the material due to metal fatigue. There is a domino effect that cuts in once a fracture has occurred. The point where the filament has narrowed due to the fracture, now has a much higher resistance and the heating at that point is much higher than the rest of the filament. Higher temperature, more flexing=failure.
If you were to inspect the motors that are failing on your RC helicopter, I am sure you will find that they are being used well outside their design specifications. It may be higher loads, more voltage, or restricted airflow. If you were to inspect an RC servo, you would find that it contains a DC motor, geared to the output shaft, and a pot for feedback all connected and driven by a PWM motor driver.
If you were to inspect and analyze the DC brushless motor that drives the fan on the CPU inside you computer, you would find a circuit that pulses the coils of the motor to make it spin.
To repeat, speed controllers and PWM drives do not damage motors. The motors supplied in our kits coupled to the speed controllers under the right design conditions will not prematurely fail. Failure of motors on FRC robots are due to entirely different circumstances not related to drive, input voltage or PWM.

Re: CIM speed in reverse?
 

I'd like to take this chance to thank Al and Mike for their contributions to this thread. Lots of great advice/knowledge being shared.


Thanks again guys, we really appreciate what you're doing. ;)

John

Re: CIM speed in reverse?
 

I've had three motors fail in the course of my RC hobbies. Once was in an RC airplane where the (non-replacable) brushes wore down really far and the motor did not spin nice anymore, lots of friction. I suspect this is what is happening in your helicopter because these motors do have a brush life on them. The second time was in an RC truck where for whatever reason the armature got bent and was slightly hitting on the magnets. The third time was in an RC buggy where something got lodged between the armature and the magents causing the motor to sieze up.

While I did have some failures, none of them were caused by pulsing.

Re: CIM speed in reverse?
 

if you look at recent high end rc escs, they have insane high frequencies. and compared to that, well analog would be the analog speed controllers. not good things. first, just plain eat up the batteries. because the resistors use the juice while escs just shut off the flow. and about those motors that last ten hours? thats alot already. 50-70? if that helicopters motor isnt just crap from the beginning, and its a as-long-as-it-spins-its-ok kind of motor, thats incredible. wish my motors would keep their perfomance up that long. my rc car motors. suppose one battery pack lasts say 5 min. id probably notice a loss in power after about at a maximum 50 packs. which is 250minutes, around 4hrs. and thats a BIG loss in efficinecy by then. the brushes would probably be almost gone by then. (im talking around 19t motors with 7.2v) then ud need to change brushes, cut comm, w/e. pros do it like every pack. are u sure the 50-70 hrs isnt with all the maintenance? cutting coms, changing brushes, reoiling the bearings/bushings? because if you ignore all that, you will finish that motor real quickly. it will stop just like sanddrags airplane's did. most good motors are openupable(sorry cant think up the right word just now) for a good reason.

Re: CIM speed in reverse?
 

Okay, I'm convinced.
Pulsing a motor, at least at the frequencies we're talking, does no harm to the motor.
Thank you for correcting me.

Sparks