Poor loco performance

Just had my LHS convert an New Old Stock Atlas GP38 (from out of the box) to DCC with a Tsunami Sound decoder. Electrically all works well, no problems. But the actual motion is jerky at low speed. Tried resetting decoder, cleaned the wheels. All other locos run fine on layout, so not dirty track. Sound and lights stay on fully, just jerky motion. LHS lubricated worm gears, axle ends, motor bearings. So before I tear into it, just looking for suggestions at to whether you think it might might be decoder or drivetrain related. Thanks all.

Really hard here, how did it run on DC? Always try to test a loco before converting to DCC, as if it’s not smooth in DC, it will not be smooth in DCC. Try to adjust the BEMF and kick start settings, and try just doing a “break in” run. The settings may help “smoothen” the running of the motor, but if it has a drivetrain issue, or a power issue, no settings will correct it.

My vote, without knowing how it ran on DC, is drivetrain/power issues. As to whether or not it will smooth out over time (break in) or is adjustable via decoder settings…

How old of old stock are we talking? Did the LHS do a thorough cleaning before the decoder install? It may even be an issue of corroded contacts somewhere inside the unit, poor solder job, etc… Lights and sound will remain on if it’s a power to the motor issue…

Download the Tsunami manual from their site. Make sure you get the correct version for the model of decoder you have. Then, look high up in the 190-212 CV range for fine motor control. Read about it, fiddle with it, and see if you can’t smoothen out that drive. I had to do this for a DC version of a J Class 4-8-4 from BLI a few years ago.

This is the detailed instruction with the author’s signature included at the end. It’s a long read, so take your time to figure out what is required, and then begin to experiment with your drive. You can always do a CV8 factory reset if things to turtle on you and start over. Do keep records, though, and do trials so that each set of CV values can be evaluated. Finally, one set will do it for you locomotive:

Tsunami PID/BEMF CV Tuning Concepts

I spent quite some time on the phone with Soundtraxx today. The following is a write up of what I was told. It includes knowledge I have gained recently in researching the PID process in general and the Tsunami decoders in particular. A PID controller is a well-known method of doing “process control”. In our case the “process” is the speed of our Tsunami equipped locomotives. And the task we are working towards is producing a process/procedure that will allow the user to find/discover the best/correct settings for their particular locomotive that will give them the performance they want.

It is also important to understand that the values for a particular locomotive – make and model and scale – may not work for ot

I had a similar problem with a Atlas GP-35. (Tsunami install I did myself). Turned out there wasn’t enough clearence in the shell which was causing the decoder to rub up against the flywheel when I had the shell fully on.

I’ll tear apart my GP38 this week and take a peek to see if there could be similar issues (I have to convert it to DCC sound any way)

Just an FYI…poor slow speed performance is not great from an out of the box tsunami. With tweek ing it can be made better, but nowhere need the quality of ESU. Lenz or QSI. In terms of tweeking, ESU has an auto calibrate with most of its newer decoders, with the LokPilot Standard being the exception. Not to mention the whole movement before loading issue with Soundtraxx. … David B

I agree with “Selector’s” comments about tweeking the higher end CV’s. You might also raise the start voltage CV 2. I did a tsunami install in an Atlas GP38 and found that out of the box the tsunami default for start voltage needed to be increased to get it going. Then changing those high end CV’s smoothed out the low speed operation. I use a straight line speed table setting and adjusted the trim settings (CV 66 and 95) to obtain the running characteristics I wanted.

Marty C

Hi,

Sometimes loco problems have pretty basic causes… You are certain the track is clean and not the source of the difficulty. But how about the loco wheels? Just for grins, put an alcohol soaked paper towel on the track and set one of the trucks on it - the other on bare track. Crank up the throttle for a few seconds, remove the loco, and look at the towel… That will tell you if the wheels were dirty or got “over lubed” by the shop.

One other thing… it is possible that the loco ran this way before the decoder was added. Perhaps there is a “rough gear” or something like that. Or, perhaps the loco just needs some serious break in time.

For what its worth…

An old Atlas “yellow box” locomotive? I have several and have put TCS T1 decoders in them and have a similar issue. They all have jerky slow speed performance.

They all ran smooth as silk on straight DC. Now, on speed steps 1-5 they creep along just great, but somewhere between 5 and 10, the start to jerk. At 10 and above, they run perfectly well.

I’ve messed with the BEMF setting some with no real improvment, although I’m not sure I’ve exhausted all the possibilities yet.

Speed step 10 is really pretty slow. I might just set that to be the minimum.

[quote user=“mobilman44”]

Hi,

Sometimes loco problems have pretty basic causes… You are certain the track is clean and not the source of the difficulty. But how about the loco wheels? Just for grins, put an alcohol soaked paper towel on the track and set one of the trucks on it - the other on bare track. Crank up the throttle for a few seconds, remove the loco, and look at the towel… That will tell you if the wheels were dirty or got “over lubed” by the shop.

One other thing… it is possible that the loco ran this way before the decoder was added. Perhaps there is a “rough gear” or something like that. Or, perhaps the loco just needs some serious break in time.

For what its worth…

thanks for the suggestions to you all. just to clarify I did clean the wheels and have been breaking it in for several days now, both directions, low and high speed. It is vintage approx. 2000 (?) Atlas Master series in black box with red printing. examining performance again last night it seems like drivetrain is binding, then frees itself. I guess I’ll have to open it up. i think it is more difficult to get the sheel off than other models.

Tophias,

On the Atlas website you can click the support tab and then go to “General Support” and you will get a drop down menu that includes “shell removal” Its a good reference to print out and save.

Marty C

Thanks Marty, that’s something I didn’t think to do. With my eyes and hands getting older I can use all the help I can get

UPDATE-finally got the shell off and found nothing to be restricting the drivetrain. Gently, slowly rotated the flywheel and found not feel at binding; no decoder wires getting in the way. All solder joints look solid. Atvfirstvwhen operating loco moves fine, but after a few minutes then the jerking begins. No issues with sound or lights flickering. Could this be something with the motor? Decoder? Would appreciate any advise you can offer. Thanks.

My suspicion would be a flaky or not properly tuned Tsunami decoder motor drive circuitry. That decoder has several CVs that can be set for smoother motor control, but if you don’t have any experience with tuning up one of them, you need to take it back to the hobby shop that installed it and see if anyone there knows how.

My first thought is that the BEMF needs to be tuned to the motor. To verify if that is indeed the case, try turning the BEMF off altogether and see if the problem goes away. If it does, you can either leave it off, or turn it back on and fiddle with the BEMF settings to smooth it out.

If the problem persists with the BEMF turned off, no point in wasting time fiddling with the BEMF settings as the problem is elsewhere …

Mark.

Thanks Mark, that seems to have had a positive effect. On JMRI I changed “BEMF CUTOUT” to a value of 1. Now runs smooth, but the prime mover notches up to like three settings before it moves (approx. 15 on my Digitrax throttle). So it now runs smoother, but I don’t see how I can put it in a consist. At least it will run by itself. Anyone have advise how to mu it? Thanks.

That is normal when turning the BEMF off, or almost off. To compensate for this, you will need to adjust the starting voltage - CV2. Keep increasing the value in CV2 until the engine just begins to move in speed step one. It should then play reasonably well with your other engines.

Mark.

THanks Mark, I’ll try that tomorrow And report back. And to Selector, if you read this, I also appreciate your input from a few days ago. It just seemed way to involved for my Old, limited capabilities. But I do appreciate the input nevertheless.

Tophias

I’m going to re-suggest what David Bedard suggested. Next time, ask for a Loksound Select decoder. Performance wise, they are so vastly superior to the Tsunami that I wouldn’t take a Soundtraxx Tsunami for free if you offered me one. I have purchased just one Tsunami, and I had to go through all the steps that Selector suggested to get it to run even half decently. I will never buy another!

The Loksound Select is also $16.00 less expensive than the Soundtraxx Tsunami (priced at Tony’s Train Exchange).

Dave

thanks Mark, that did it. Spent quite some time fiddling with JMRI settings, ending with BEMF Cutoff at 5, V-Start at 42, and Accel and Decel at 50. It actually plays rather nicely with its mate now. I thank you for helping me salvage this project. And one final question for my general knowledge-I had this loco converted to DCC along with an identical mate. Same locos (both new, old stock, purchased at same time) with same decoder Installed. Why this large discrepancy in performance? Thanks for everyone’s input, much appreciated.

Best I can figure is just variations in motor characteristics. These aren’t military grade by no means. Just a variation in the spring pressure of the brushes will affect motor performance.

I had six brand new engines that I installed identical decoders in, and he also wanted them speed matched. None of the settings for each engine were the same. I was able to get them to run identical to each other, but there was no way I could use the same settings for each one.

As the engines get a number of miles on them and things begin to get run in, no doubt these settings will have to be tweaked on occasion to keep them all in perfect harmony.

Mark.