Hello not sure if this thread belongs to this forum or large scale but posting here

Recently I got a Piko 38237 mini American Mogul locomotive it comes with soundtraxx DCC decoder installed at factory. I want to run this on my relatively small garden layout. I use CS3 to run LGB locomotives and accessories on this layout without any issues. The track has many contact detectors via S88 for automation. I have the Gauge 1 marklin 60195 power supply as well as a 19V 2A Marklin HO power supply for my HO layouts.

when I first attempted to run Piko I had lots of difficulty with stutter and lights flashing, it would refuse to work smoothly. I contacted Piko support and they pointed to voltage being too high for this locomotive (cannot handle more than 27V and will flash lights to warn the user).... I switched to the 19V HO power supply (limited to 2A so can't run many trains with this but up two locomotives is OK) and the problem got better but I encountered the same issue somewhat less frequently....

So I decided to bring an oscilloscope and measure the track waveform methodically. I observed something unexpected.

When the PIKO 38237 locomotive is stationary on the track with sound, light and smoke functions ON, the track waveform is normal and precisely measures -18V to + 18V square (36V peak-to-peak) which is NMRA compliant and the same voltage reported by CS3. Now If I start the motor I see quite a bit of over-shoot on the waveform it looks very noisy and the peak to peak can increase dramatically as the locomotive goes around a loop. This happens even at slow speeds. I verified when stuttering happens the waveform peak to peak exceeds 54V! This is not normal behavior as the track current draw is < 500mA (power supply rated by 2A) and CS3 shows no overload warning.

Then I proceeded to measure several LGB locomotives to see if they also show the same behavior, I tested the heavy LGB 28442 Ge 4/4 II locomotive. This is a dual motor large lcomcotive and draws a large current. Even at very high speed the waveform is stable with peak to peak voltage steady a 36V… does not show “ noise” like the Piko… I tested two other LGB locomotives (one steam and one diesel). The waveform is clean even when all of them are running at the same time…. The waveform is clean when Piko is stationary and multiple LGB locomotives are running on the same track. The abnormal waveform happens only when Piko motor is running. Even when the Piko does not stutter the abnormal waveform is seen but with ~40-45V peak-to-peak voltage


I verified this has nothing to do with MFX, I disabled all protocols on the CS3 except for DCC, disconnected all accessories and rebooted but observed exactly same behavior. It seems the decoder is not drawing current in a smooth load balanced way… causing this instability on track waveform.

has anyone seen this issue with Piko/soundtrax locomotives ?

Fig 1 track waveform with Piko ON but stationary. +18V to -18V square wave DCC 36V peak to peak



Fig 2 track waveform with Piko running at 10% speed peak to peak goes to 43V


Fig 3 track waveform when piko stutters at 10% speed, lights flashing on the Piko peak to peak at 54V


Fig 4 track waveform with LGB 28442 at 80% speed, all is normal peak to peak stable 36V


Fig. 5 LGB and Piko on the same tracks