Tuning a Massoth LS and L to run together

andyspencer

andyspencer

Registered
Country flag
United-Kingdom
Hi all, current little issue I've got - trying to get a Massoth L and LS decoder to run together. They both have the same speed curve settings and start voltages etc, but the L wants to pull away slightly before the LS and the LS doesn't want to stop as quickly. Am I missing something silly? Anybody else had any joy making these to decoders behave together?
 
the motors of 2 locos are never exactly the same.

adjust CV2 in one of the locos to match start speeds.

on stopping, why not set CV's 3 and 4 to zero in both and see if that makes them match up better... if you want them to stop quickly, then you need to use a braking function or set acceleration and deceleration to zero.

Greg
 
Andy, you haven't got one old, and one new, motor have you?

You may have to swap a few motors about, to find two with similar characteristics?? :think:
 
Greg Elmassian said:
I'm not wanting to be confrontative, but why would you want to go to all the trouble of swapping (with the distinct possibility of just swapping the problem between locos) when DCC has the capability to match the locos, with software settings?

Greg
Click to expand...

IF a new Buhler motor has been fitted to one of the units, then they will never match..
Design (internally) has changed, making the new motors more-efficient. - Starting voltage, and back-emf are way off between old and new.
I suppose you might be able to fudge it, with custom speed-curves, but it would only be good for light-engine (say).. Put a load on it, and the differences in back-emf would kick-in, and back to square one.
 
Really?

So, let's ignore the BEMF issue first, and then bring it in later.

I cannot believe that a custom speed curve cannot match the 2 locos running light. I have set locos that will run for hours exactly 4 feet apart.

Depending on the decoder, you may or may not need CV2 and CV5... most times the speed curve alone is all that should be set.

Then, some decoders have a "kickstart", a little extra kick to overcome initial friction.

Now, before we get to BEMF... one of the common misconceptions is that you need 2 locos running exactly the same, so people check them with no loads and not coupled together, that seems logical. But when you couple 2 locos, if there are not massive differences, they equalize out... when a load is applied things change a bit, and the "faster" loco will draw more current as it works harder, and the slower loco, being more lightly loaded, will go faster... so the speed, when loaded, seems to equalize much better than people think.

Now let's add in BEMF... BEMF is not simple, because it has 3 parameters and can be "tuned" to have very different effects. One extreme of tuning is to make the loco on "cruise control" where the speed is independent of load... nice for a single loco running a train with a wide variation in grades... you can set it at a speed and know it will get up the grade and not run away down grade.

But the main reason for development is to smooth out the performance of the loco under irregular conditions, short term, like a binding spot, "stiction" when starting, etc. Tuned this way, I have no issues consisting multiple locos, although the "common wisdom" is to turn BEMF off when consisting.

So my advice is turn off BEMF so as not to be confusing at first, pick one loco as the "standard" and set the speed table in the other to match. I do this with POM in real time... I put something in the couplers so the locos don't lock together.

(I actually now set all locos with a custom speed table vs. a scale speedometer, mapping speed steps to SMPH so all locos are matched to each other)

Greg
 
Sometimes I feel that in 12 years of running G scale I know absolutely nothing..... Oh no it is not that I know nothing...it is just that some folk know a lot..... and they like making sure that you know that fact. :eek::think:;);)
 
Yeah, makes my brain hurt - good ol'e auntielog suits me, one black wire, one red wire :smoke::smoke::smoke::smoke:
 
Well, I know that is probably not a friendly remark from my friend Mike, but I'm presenting a few years of experience here, and the compendium of what I have learned and executed to try to give the OP an answer.

If there's things wrong in my statements, please by all means, let's discuss, my post is to try to answer the OP's question, and I consist locos about 90% of the time, in G and Z scales.

There's a lot of "common knowledge" that is rooted in history, and some is no longer true. I started G scale when I was told that DCC outside was flat out impossible.

I have been told that BEMF is never to be used in consisting, I have been told a lot of things.

But I have a scientific background, so I always ask for an explanation when things don't make sense... often there is no explanation, but just something someone heard 10 years ago..

Truly, 10 years ago DCC in large scale was iffy, because there was no sense to prevent the microprocessor from brief power interruptions... now virtually all DCC decoders have some on-board capacitors to keep the micro "alive"...

When someone starts talking about BEMF just being an on/off situation, which is true on some decoders, it does not tell the whole story. There are the 3 parameters (PID) that can flavor BEMF quite a bit, as I tried to convey.

All I am trying to convey is that there are definitely alternatives to replacing motors in DCC locos to match 2 locos... and it's not too tough.....

If it's not deemed helpful, just let me know.. I was providing reasons and facts to back up my statements, not just "oh you are wrong about BEMF" or "you are wrong about speed tables" p.s. this is the DCC forum, and a DCC question, sorry it is not as simple as analog.

So, apparently I am damned if I do (provide explanations why)

and damned if I don't (not provide explanations/backup for my statements)


This is a complex issue, and not everyone is familiar how the stuff works together, I took the OP's request for help literally... noted the statement was "if xyz, then they will never match"... I disagree and provided support for my statement. This was for the benefit for the OP...
 
Greg Elmassian said:
Well, I know that is probably not a friendly remark, but I'm presenting a few years of experience here, and the compendium of what I have learned and executed.

If there's things wrong in my statements, please by all means, let's discuss, my post is to try to answer the OP's question, and I consist locos about 90% of the time, in G and Z scales.

There's a lot of "common knowledge" that is rooted in history, and some is no longer true. I started G scale when I was told that DCC outside was flat out impossible.

I have been told that BEMF is never to be used in consisting, I have been told a lot of things.

But I have a scientific background, so I always ask for an explanation when things don't make sense... often there is no explanation, but just something someone heard 10 years ago..

Truly, 10 years ago DCC in large scale was iffy, because there was no sense to prevent the microprocessor from brief power interruptions... now virtually all DCC decoders have some on-board capacitors to keep the micro "alive"...

When someone starts talking about BEMF just being an on/off situation, which is true on some decoders, it does not tell the whole story. There are the 3 parameters (PID) that can flavor BEMF quite a bit, as I tried to convey.

All I am trying to convey is that there are definitely alternatives to replacing motors in DCC locos to match 2 locos... and it's not too tough.....

If it's not deemed helpful, just let me know.. I was providing reasons and facts to back up my statements, not just "oh you are wrong about BEMF" or "you are wrong about speed tables"

So, apparently I am damned if I do (provide explanations why)

and damned if I don't (not provide explanations for my statements)


This is a complex issue, and not everyone is familiar how the stuff works together, I took the OP's request for help literally... noted the statement was "if xyz, then they will never match"... I disagree and provided support for my statement. This was for the benefit for the OP...
Click to expand...
No, no - it's just way above my pay grade o_Oo_O

I have a brain that struggles with things that can't be seen - I manage OK with simple electrics, I understand the three-phase stuff, but can't get my head around why it does what it does.

So I stick with where I'm happiest, one brick on another :rock::rock::rock:
 
It does not help that different manufacturers do things differently either... if there were standardized ways of setting BEMF, or other features it would probably help.

Eventually someone will probably do something at a bit higher level and simplify things... JMRI has some potential, since it can adapt to different manufacturer's stuff.

I have seen a setup where JMRI controls and programs the loco, and also reads from a trackside speedometer, and basically speed matching becomes an automated process.

That would be a boon to many people.

Programming lighting functions, perhaps with a camera looking at the loco, detecting the lighting is another thing that could be done soon.

The bane and boon of DCC is that it is so programmable... so many different things to be "tweaked"... I have a friend that wanted a loco to sound like the one he remembered. Well, with 32 different sounds and 2 speakers and a 32 channel "mixer" for volume and balance, it took the better part of a day to get it the way he wanted... he's happy but I was tired...

Greg
 
Being a lot closer to Andy.. And having had a telephone conversation over the original problem..

It would appear the main problem is the two decoders are not responding in the same way. - The LS seems to have momentum settings 'on' and set so that motor starts later, and stops later, than the motor controlled by the L decoder.

My test was to get the motor-blocks swapped between the two decoders. - Whichever motor is on the LS responds in the same way, as if the momentum settings are on, and a little high.

It is beginning to look like either the decoder is not happy with the CV settings it has been given, is not acting on said instructions, or there is a hardware/firmware fault with the LS decoder.
 
Greg Elmassian said:
does he not have the capability to program CV's 2,3,4,5? My first suggestion and first reply to his post.
Click to expand...

Yes, set 'same and low' on both.. - The fact the LS will not 'take' these changes (they read back as the programmed values, but there is still delay in starting and stopping) suggests a decoder problem.

Greg Elmassian said:
Also, not knowing what system
Click to expand...

LS and L decoders..
All Massoth equipment, though there is the option of JRMI to read the decoders..
 
I have noticed some decoders have a self-imposed delay when reversing directions even with CV 3 and 4 set to zero.

Do you think that an incorrectly set CV2 could be interpreted as a different momentum? ... I've certainly seen this happen and be interpreted as momentum.

I would suggest you adjust CV2 on each loco to just start at speed step 1? (or just set them so the both start at the same speed step if you cannot get them to start that slowly)

Also are you running 14 or 28 or 128 speed steps? I prefer the higher resolution, but some people don't like it... it takes too much knob twirling to go fast... in any case, run them both as the same setting.

If it was me, again, I would set the 2 up so their starting speeds CV2, and top speeds CV5 are matched, from the perspective of the locomotive performance, not the CV values of course... also, you did not mention, but have you verified firmware versions same?

Greg
 
I am thinking that either Andy wait for a replacement LS,
Or,
There were some LS decoders out there with a 'read-only' bit set.. - I had one of these, and was told by Massoth to load an XLS decoder image, which worked!
 
You must log in or register to reply here.
Back
Top Bottom