[SOLVED] Noisy motors on LGB two motor Bo-Bo diesel 23510

idlemarvel

idlemarvel

Neither idle nor a marvel
13 Jul 2015
2,103
278
Ascot
Just bought one of these 2nd hand via eBay from Rails of Sheffield. The model is immaculate, hardly a fingerprint. From the factory these are analog with 6-pole DCC interface. The previous owner had installed a decoder (decoders) unspecified on address 8. When I opened it up I found this ball of spaghetti:

IMG_20200521_152548.jpg


There are two LGB 55020 decoders, one attached to the track and motor pins and the 6-pole interface (that rainbow coloured ribbon cable), and the other just to the other set of track and motor pins. They work (loco moves, lights come on) but it sounds like this:


What on earth? I assume it is the decoders forcing the motors to make this racket. AFAIK back EMF is off (CV54 = 0) but does anyone have any other suggestions?

I have an analog controller so I'm going to remove the decoders, set the DIP switches back to analog and see if it makes the same racket in analog mode. I hope not. If they are quiet in analog mode I will replace the 2 x 55020 with an XL.

I hope I can fix this as otherwise the loco is as good as new.
 
dunnyrail

dunnyrail

DOGS, Garden Railways, Steam Trains, Jive Dancing,
25 Oct 2009
16,905
1,542
72
St.Neots Cambridgeshire UK
Just bought one of these 2nd hand via eBay from Rails of Sheffield. The model is immaculate, hardly a fingerprint. From the factory these are analog with 6-pole DCC interface. The previous owner had installed a decoder (decoders) unspecified on address 8. When I opened it up I found this ball of spaghetti:

View attachment 266722

There are two LGB 55020 decoders, one attached to the track and motor pins and the 6-pole interface (that rainbow coloured ribbon cable), and the other just to the other set of track and motor pins. They work (loco moves, lights come on) but it sounds like this:


What on earth? I assume it is the decoders forcing the motors to make this racket. AFAIK back EMF is off (CV54 = 0) but does anyone have any other suggestions?

I have an analog controller so I'm going to remove the decoders, set the DIP switches back to analog and see if it makes the same racket in analog mode. I hope not. If they are quiet in analog mode I will replace the 2 x 55020 with an XL.

I hope I can fix this as otherwise the loco is as good as new.
Can you do another pic from the other side with the spag bol untangled a little. My feeling is that the chips should have been connected to 4 pins each, but it depends on the interface and perhaps the 6 way is correct? Also what does the symbol underneath show, and what is the loco LGB Reference number. That is also likely to be u derneath but can also in some cases be printed on one of the weights.
 
idlemarvel

idlemarvel

Neither idle nor a marvel
13 Jul 2015
2,103
278
Ascot
I should have added that I checked the motor frequency CV 9 and it was set to 128 which AFAIK is not a valid value. I set it to 0 which is 16KHz but it made no difference to the noise.
I assume that both decoders get updated with CV values as it read back the changed value okay.
The starting voltage CV2 was set to 0 which was a bit weird, I reset it to 8.
 
ntpntpntp

ntpntpntp

Registered
24 Oct 2009
7,297
52
57
UK
idlemarvel

idlemarvel

Neither idle nor a marvel
13 Jul 2015
2,103
278
Ascot
idlemarvel

idlemarvel

Neither idle nor a marvel
13 Jul 2015
2,103
278
Ascot
Can you do another pic from the other side with the spag bol untangled a little. My feeling is that the chips should have been connected to 4 pins each, but it depends on the interface and perhaps the 6 way is correct? Also what does the symbol underneath show, and what is the loco LGB Reference number. That is also likely to be u derneath but can also in some cases be printed on one of the weights.
The chips are connected to 4 pins each, and in addition one of them is connected to the 6-pin interface, to drive the lights as well as one of the motors I assume. The loco is a 23510, the symbol is the 6 pin decoder interface symbol. I think ntpntpntp ntpntpntp has got the probable answer.
 
idlemarvel

idlemarvel

Neither idle nor a marvel
13 Jul 2015
2,103
278
Ascot
Completed analog test. Motors smooth as silk and quiet as a mouse.

Those 55020 are truly cr@ppy. I'll keep them as coach lighting decoders and replace them with an XL.

Thanks dunnyrail dunnyrail and top marks to ntpntpntp ntpntpntp

I love this forum, answers within an hour of asking... :h:
 
PhilP

PhilP

G Scale, 7/8th's, Electronics
5 Jun 2013
25,513
1,316
Tamworth, Staffs.
The 55020 is actually a Lenz decoder, and certainly by today's standards, 'crappy' probably sums it up! :nerd:
 
-bbbb

-bbbb

Registered
21 Dec 2017
318
84
Idaho
PhilP

PhilP

G Scale, 7/8th's, Electronics
5 Jun 2013
25,513
1,316
Tamworth, Staffs.
If there is low frequency PWM, and the motor hums like this, is that a problem for the motor, or is the noise the only problem?
That depends on the motor characteristics, mainly..

If the motor is making a noise, it is usually the windings are buzzing/vibrating. This can not be good for any fine-wire-joints, and can also generate heat, in extremis causing the motor windings to swell.

Most modern motors are designed to run at higher PWM frequencies, and may not like a low frequency.

If you can stand the noise, then fine..
I would run a loco for a few minutes, then see if the motor was appreciably warm.. If it was, I might run for a few more minutes, to see if it gets hot.

I have had a situation where a motor would get too hot to hold, after running for 2-3 minutes, with a PWM frequency of 2kHz. - Run the same motor at 16kHz, and it is only warm after a few hours running.
 
P

Paul M

Registered
25 Oct 2016
4,570
729
57
Royston
If there is low frequency PWM, and the motor hums like this, is that a problem for the motor, or is the noise the only problem?
If any motor makes a noise it shouldn't, it isn't happy.
 
dunnyrail

dunnyrail

DOGS, Garden Railways, Steam Trains, Jive Dancing,
25 Oct 2009
16,905
1,542
72
St.Neots Cambridgeshire UK
The chips are connected to 4 pins each, and in addition one of them is connected to the 6-pin interface, to drive the lights as well as one of the motors I assume. The loco is a 23510, the symbol is the 6 pin decoder interface symbol. I think ntpntpntp ntpntpntp has got the probable answer.
Ok thanks for that, perhaps a way forward may be to unwire each bogie then try the remaining one disconnected on blocks then try the same with the other to see if it is just 1 motor or both being noisy. Make sure you remember which pins are disconnected before you do so that they can go back the right way. Picture of the pins connected before you start will help you ensure things go back together the correct way. Each bogie should have a 4 pin plug to it as well that will need to be disconnected one at a time. They normally have 4 colours to the bogie but as I cannot see from your picture look for 2 sets of 4 wires disappearing through the chassis, they will be to the bogies.
 
idlemarvel

idlemarvel

Neither idle nor a marvel
13 Jul 2015
2,103
278
Ascot
To wrap this up, here is a video clip of the same loco with a Massoth XL instead of the 2 x LGB 55020.
As you can hear, almost zero noise and smooth start. Default settings on the decoder.
So the problem was with the old 55020 decoders, thanks for all the other thoughts, but no need to explore further.
I may start another thread about how to install an XL on this loco.
 
dunnyrail

dunnyrail

DOGS, Garden Railways, Steam Trains, Jive Dancing,
25 Oct 2009
16,905
1,542
72
St.Neots Cambridgeshire UK
To wrap this up, here is a video clip of the same loco with a Massoth XL instead of the 2 x LGB 55020.
As you can hear, almost zero noise and smooth start. Default settings on the decoder.
So the problem was with the old 55020 decoders, thanks for all the other thoughts, but no need to explore further.
I may start another thread about how to install an XL on this loco.
Result.
 
-bbbb

-bbbb

Registered
21 Dec 2017
318
84
Idaho
When I read about motor noise being reduced by capacitors added to the motors, should I take that as a different kind of motor noise, or could capacitors help with pwm problems?
 
Greg Elmassian

Greg Elmassian

Registered
8 Mar 2014
3,761
485
San Diego
www.elmassian.com
Different... the capacitors reduce the electrical interference to OTHER devices from the motor.... and this is normally electromagnetic interference.

This thread is on audible noise from the motor from a poorly designed electrical circuit in the decoder.
 
D

Dan

Registered
28 Jan 2010
290
20
Eastern MA
I agree that the 55020 decoders are not very good and the LENZ version has the LENZ logo etched on the board. these were very limited and only had a few CV's that could be changed. 55020 with no logo were better, more cv's to change but still had limitations like no 128 speed steps. 55021 were a step in the right direction but only latest versions could do 128 speed steps. For the money, there are much better decoders for less $$ then any of these.