ESU 5XL Sound Decoder Install - Motor Hum Noise

[LGB333]

Hi Fellow Hobbyists - I just completed a 5XL sound decoder installation into an LGB 28002 Rugen Steamer along with a Massoth Pulsed Smoker. Everything works great except I notice a slight motor hum noise when I operate the loco on my test stand when the the sound is off. I set the Motor Control for Linear and Buhler/Kiss Motors which has worked well for all the other 5XL installs I've done on LGB locomotives. I tried turning off the Load Control/EMF but then the motor won't operate so that's not an option to eliminate the motor hum noise. What other options should I try to remove the motor hum noise?

 

[idlemarvel]

My first thought would be the hum is due to back EMF. When you tried turning that off did it eliminate the noise? If it did then next question is why did turning BEMF off cause the motor to not operate. Did you increase the start voltage CV2 to compensate for no BEMF?

 

[LGB333]

My first thought would be the hum is due to back EMF. When you tried turning that off did it eliminate the noise? If it did then next question is why did turning BEMF off cause the motor to not operate. Did you increase the start voltage CV2 to compensate for no BEMF?

Hi Dave - Okay, I turned off the Load Control/EMF again and there's no motor hum. I hadn't increased the CV2 Starting Voltage high enough when I tested it earlier. This time I keep increasing CV2 (originally 2) from 20 - 70 and got the motor to start moving at step 1.........runs good with no hum noise. But when I turned on the sound, the electronic chuff is now way too slow about 1 chuff every 4 wheel revolutions.....I can adjust that.

I've installed ESU 5XL sound decoders into about 12 - 15 LGB locomotives over the last 12 months using the standard Linear and Buhler/Kiss Motor settings and never had a motor him noise on any of them. I'm wondering why this particular LGB 28002 locomotive's motor is causing the hum noise. These were produced by LGB during 2000 - 2002 timeframe and the locomotive owner has not used it very much so it's in excellent condition. It's very heavy with two large lead weights. I checked the LGB User Guide and it states the motor is #62210 28001- series motor. All the other LGB locomotives I've installed ESU 5XL decoders had either the regular short shaft or long shaft motors, so maybe it's an issue with this particular style motor.

 

[LGB333]

Hi Dave - Okay, I turned off the Load Control/EMF again and there's no motor hum. I hadn't increased the CV2 Starting Voltage high enough when I tested it earlier. This time I keep increasing CV2 (originally 2) from 20 - 70 and got the motor to start moving at step 1.........runs good with no hum noise. But when I turned on the sound, the electronic chuff is now way too slow about 1 chuff every 4 wheel revolutions.....I can adjust that.

I've installed ESU 5XL sound decoders into about 12 - 15 LGB locomotives over the last 12 months using the standard Linear and Buhler/Kiss Motor settings and never had a motor him noise on any of them. I'm wondering why this particular LGB 28002 locomotive's motor is causing the hum noise. These were produced by LGB during 2000 - 2002 timeframe and the locomotive owner has not used it very much so it's in excellent condition. It's very heavy with two large lead weights. I checked the LGB User Guide and it states the motor is #62210 28001- series motor. All the other LGB locomotives I've installed ESU 5XL decoders had either the regular short shaft or long shaft motors, so maybe it's an issue with this particular style motor.

Well, the locomotive's operation isn't that good with the Load Control/EMF OFF. So I'm going to keep the decoder with Load Control/EMF ON and live with the slight hum noise from the motor. The hum seems less pronounced now, too, with the Load Control/EMF ON. And you don't notice the slight motor hum anyway when the decoder's sound is on.

 

[phils2um]

Tom, did you try using the ESU decoder's motor self-calibration feature? That may get you better starting values for the motor CV settings. All you need is a long enough straight or a R3 minimum (so the loco doesn't fly off the track!) circle of track. Don't remember how to do it but the procedure is explained in the decoder manual.

 

[Greg Elmassian]

Does the self calibration feature also change the PWM frequency? Or is it just the BEMF PID parameters?

Greg

 

[LGB333]

Tom, did you try using the ESU decoder's motor self-calibration feature? That may get you better starting values for the motor CV settings. All you need is a long enough straight or a R3 minimum (so the loco doesn't fly off the track!) circle of track. Don't remember how to do it but the procedure is explained in the decoder manual.

Hi Phil - Yes, I also tried that. You set CV54=0 and then press F1. I use my Massoth roller wheels test stand for this procedure instead of a streach of track. Results: Same slight hum noise. The CV9 Frequency standard setting when using the Linear and Buhler/Kiss Motor settings equals 40kHz. For comparison, Massoth XLS Configuraton Manual indicates 16kHz is the optimum for motor operation. So on the ESU 5XL I changed CV9=16 (16kHz) and the motor seems quieter at this frequency. So, unless someone has a better fix, I think I'm going to keep the Linear and Buhler/Kiss motor control settings, and also use CV2=3 and CV9=16.

 

[Greg Elmassian]

You need to have it run on the track! It's trying to sense the bemf needed, and the motion of the loco and it's momentum is important.

Very weird that the motor is quieter at 16 KHz, that is audible to most people.

Greg

 

[LGB333]

Does the self calibration feature also change the PWM frequency? Or is it just the BEMF PID parameters?

Greg

Greg - I don't think it changes the PWM frequency which is 40kHz when I use the Motor Control settings of Linear and Buhler/Kiss motor. The ESU 5 Users Instructions state that the Auto Calibration of the Motor "........the identified load compensation parameters will be saved in the CV51, 52, 53, 54, and 55. As I indicated in my reply to Phil, I changed PWM Frequency CV9 from 40 to 16 which does reduce the noise some. By the way, someone told me that the standard LGB short shaft and long shaft motors operate at 14 volts whereas these unique LGB 62210 28001-series motors operate at 22 volts and are only installed in LGB 28000 series Rugen locomotives. All my other ESU 5XL decoder installs have been in LGB locomotives with the standard Buhler motors and this is the first time I've encountered the motor noise issue.

 

[LGB333]

You need to have it run on the track! It's trying to sense the bemf needed, and the motion of the loco and it's momentum is important.

Very weird that the motor is quieter at 16 KHz, that is audible to most people.

Greg

Okay, better to do it on the track. Maybe I'll do it again on the track tomorrow. But after doing it on the roller wheels test bed, it didn't reduce the motor noise any.

Regarding 16kHz, for comparison, Massoth eMotion XLS sound decoder documentation states "at 16kHz the motor is operated at its optimum and runs very quietly."

 

[Greg Elmassian]

I did not make myself very clear, sorry. The BEMF calibration should be much better using the mass of the loco. The PID parameters need something to "work against"

I don't think the BEMF calibration will help the noise.

I would also check to see if this motor has noise caps and/or chokes, and remove them.

On the statement about "runs very quietly", BS really BS.. 16 KHz can be hear by most humans... the person writing that statement must be hearing impaired compared to a young person.

Up until my 40-50's I could hear 20 KHz... 16 is irritating... that is old technology, old switching supplies hummed also until they moved their PWM frequency to ultrasonic...

Greg

 

[PhilP]

Hum suggests an ailing motor to me? - too much current being pulled, and windings moving??

I would also tease out the wiring, in case it is pickup/interference in the loom.

PhilP

 

[phils2um]

By the way, someone told me that the standard LGB short shaft and long shaft motors operate at 14 volts whereas these unique LGB 62210 28001-series motors operate at 22 volts and are only installed in LGB 28000 series Rugen locomotives.

Most certainly an urban myth aka old wives tale. The old LGB analog locos did not have voltage limiters in them or operate at a fixed voltage. The 5006/5007 transformer/analog controller could provide up to 21VDC to the motors! Varying the voltage changes the speed.

 

[wandgrudd]

All,


Could it not just be because the Rugen uses a different motor, and thus needs slightly different settings, remember virtually off the other lgb locos use the same motor but with just a short or long shaft, were as the design and size of the rugen motor is completely different and is exclusive to that series of locos?

 

[idlemarvel]

I had a Rugen 0-8-0 and I noticed a hum as well. I think the 4 fixed axle motor block which I believe is unique to this loco causes more drag on the single motor than most other locos. It definitely needed BEMF on to cope with R1 curves otherwise it just stalled. Maybe this is something to do with it.

 

[LGB333]

I did not make myself very clear, sorry. The BEMF calibration should be much better using the mass of the loco. The PID parameters need something to "work against"

I don't think the BEMF calibration will help the noise.

I would also check to see if this motor has noise caps and/or chokes, and remove them.

On the statement about "runs very quietly", BS really BS.. 16 KHz can be hear by most humans... the person writing that statement must be hearing impaired compared to a young person.

Up until my 40-50's I could hear 20 KHz... 16 is irritating... that is old technology, old switching supplies hummed also until they moved their PWM frequency to ultrasonic...

Greg

Greg etal - I'm not familiar with motor noise caps and/or chokes.........what do they look like and where would they be? I removed the factory circuit board to install the ESU 5XL decoder and I don't see any other devices around the motor although I only removed the bottom cover to add some new grease and silicone lubricant to the axles, so maybe I missed them?

The locomotive operates well with just using the Linear and Buhler/Kiss motor settings on the ESU LokProgrammer. So, if using the Auto Calibration of the Motor feature won't improve the motor noise, the only issue I'm interested in correcting, then I don't think I need to do the "on the track" Auto Calibration procedure. I was going to leave CV9=16 for 16kHz PWM frequency based upon the statement in the Massoth documentation for their XLS sound decoders, but if you recommend putting it back to the original 40kHz I'll do so. Like Phil, I'm thinking maybe this old motor installed at the LGB factory in the 2000 - 2002 timeframe, even though not used much, could be acting up with the noise. But again, the locomotive operates well, and doe not make the noise when I turn OFF the Load Control/EMF setting on the LokProgrammer. And another data point, Dave Miller indicates he had one of these LGB Rugen locos with the unique motor and found the Load Control/EMF needed to be ON for it to operate well.

So, here's my proposed course of action:
1. Keep the LokProgrammer's Load Control/EMF setting ON;
2. Use PWM frequency 16kHz or 40kHz depending on your advice;
3. Live with the motor's slight hum noise which can't be heard anyway when the sound system is operating;
4. Replace the motor down the pike when/if it fails.

Thoughts?

 

[Greg Elmassian]

The noise suppression stuff is normally mounted right on the motor, a little surprised you have not seen this, Bachmann often has both caps and inductors.

The signature of having these components is that it confused the BEMF, and that is exactly what you are telling us, that turning BEMF off stops the noise.

Also, don't understand at all your reticence to calibrate BEMF properly, since again you say turning BEMF off stops the noise.

You are confusing PWM frequency whine with BEMF "buzzing" or confusion.... sorry, again I thought that you had run into these before.

Both can make noise.

Greg

 

[LGB333]

The noise suppression stuff is normally mounted right on the motor, a little surprised you have not seen this, Bachmann often has both caps and inductors.

The signature of having these components is that it confused the BEMF, and that is exactly what you are telling us, that turning BEMF off stops the noise.

Also, don't understand at all your reticence to calibrate BEMF properly, since again you say turning BEMF off stops the noise.

You are confusing PWM frequency whine with BEMF "buzzing" or confusion.... sorry, again I thought that you had run into these before.

Both can make noise.

Greg

Greg - I'm not reticent to calibrate the BEMF properly. Unless I'm missing something, the LokProgrammer is doing that by using the following motor settings for LGB motors: Linear; Load Control/EMF ON; and Buhler/Kiss motors. So, if by "properly" you mean using the Auto Calibration of the Motor feature, it's not needed since these LokProgrammer settings do to the trick........the motor operates great but has the slight buzz noise. So maybe it has caps or inductors on the motor as you state.

Okay, last question, what PWM frequency do you recommend I use on this motor? The LokProgrammer set it at 40kHz..........do you recommend using that instead of the 16kHz that Massoth still touts in its decoder documentation for optimum operation.

 

[phils2um]

Hi Tom,

With regard to Massoth's BEMF PWM frequency. Massoth recommends 16 kHz because of the four options their XLS decoders have, 60 Hz, 250Hz, 2 kHz, or 16 kHz, it is by far the best choice for LGB motors. To quote the Massoth manual "Basically the motor runs quieter and smoother when the control frequency is increased." This does not necessarily mean 16 kHz is the optimum PWM frequency, If the Massoth decoders offered 40 kHz as a PWM frequency my guess is they would recommend it over 16 kHz but perhaps not.

The following is quoted from the ESU V5 decoder manual:
"11.3. Adapting Load Control Frequency
In the LokSound decoders load control works with 40 kHz normally. Sometimes it can be useful to reduce this frequency by half: For motors with little power due to high inductivity. If suppressors (such as capacitors, chokes, etc.) disturb load control but cannot be removed (e.g. some older Gützold® locos). The PWM frequency can be set in CV 9 as a multiple of 1000 Hz. The range of 10 to 50 allows frequencies from 10 kHz to 50 kHz."

By the way, I've looked at the motors of many LGB locos in the course of adding decoders/maintaining my fleet dating from early 1980 Stainz to current MLGB production. I can't recall ever seeing noise suppression caps or chokes on the motors. That's not to say they may not exist within the motor shell itself but that would be unusual. I admit that I've never pulled apart one of the Beuhler motors so don't know. Perhaps others can chime in with their experience.

 

[Greg Elmassian]

Yes, auto cal... notice you say the loKProgrammer settings are fine, but the motor still buzzes... So if it was me, I would try the auto cal, in case it does better, I really don't see how you can assume the stock settings are the best... and how you can be sure they have nothing to do with the buzzing.

Again, turning BEMF off stops the buzzing, right? Auto cal affects BEMF settings, right? Your statement says they have nothing to do with each other.

I did make a recommendation on PWM frequency, something ultrasonic... which is normally defined as over 20 or 21 KHz....

I don't know what the decoder can do but I can give you a recommendation that is normally the most quiet for any motor... maybe not the best performance for all motors.

on the supressor circuitry, I have never found it inside the motor.

Greg