rrinker It takes the CVs, they just don't work exactly the same as other decoders. Check CV2 and see what your real starting speed is set to. I've never messed around with speed tables, never felt the need - but then I only use decoders that support CV2, 5, and 5 for simple 3 step tables, which gets me the starting speed I want, the top speed limit I want, and the throttle response I want based on CV6. 1.3 SMPH starting speed is pretty decent. Also set some starting momentum in CV3. --Randy
It takes the CVs, they just don't work exactly the same as other decoders. Check CV2 and see what your real starting speed is set to.
I've never messed around with speed tables, never felt the need - but then I only use decoders that support CV2, 5, and 5 for simple 3 step tables, which gets me the starting speed I want, the top speed limit I want, and the throttle response I want based on CV6. 1.3 SMPH starting speed is pretty decent. Also set some starting momentum in CV3.
--Randy
Randy (and Mark),
Doing a bit of studying, I find there's a "bit 4" option that is entered when setting CV29. If "bit 4" is set to "off", one can use CV2, CV6, and CV5. If "bit 4" is set to "on", CV67-CV94 (speed table) is enabled. I believe, in the latter case, you also have to activate the speed table using CV25. I read that it should be set a a value of 16 for this. I do wonder where that number came from. If CV25 were only off-or-on, it would seem the CV25 options should be 0 or 1.
But. It appears you can only do one or the other. If you are using CV2, for example, you cannot be using the speed tables. And if you are using the speed tables, you cannot be using CV2.
So it appears to me.
Ed
gdelmoro tstage Ed, Do you really want to mess with speed tables? You might try setting CVs 2, 5, and 6 first to see if you can achieve what you want. It's actually quicker than speed tables and I've gotten very good results. You might also try running the "automatic motor turning feature" after setting your speed CVs: Place locomotive on minimum 3' piece of track Change CV54 to a value of "3" "0" Press F1 The locomotive will start out at full throttle but will quickly come to a stop. I did this for a Loksound Select decoder in a brass H20-44 and was getting well under 1sMPH at speed step 001. Tom This CV 54 thing is very interesting. where can i find out more about it? Do you do this with all your locomotives?
tstage Ed, Do you really want to mess with speed tables? You might try setting CVs 2, 5, and 6 first to see if you can achieve what you want. It's actually quicker than speed tables and I've gotten very good results. You might also try running the "automatic motor turning feature" after setting your speed CVs: Place locomotive on minimum 3' piece of track Change CV54 to a value of "3" "0" Press F1 The locomotive will start out at full throttle but will quickly come to a stop. I did this for a Loksound Select decoder in a brass H20-44 and was getting well under 1sMPH at speed step 001. Tom
Ed,
Do you really want to mess with speed tables? You might try setting CVs 2, 5, and 6 first to see if you can achieve what you want. It's actually quicker than speed tables and I've gotten very good results.
You might also try running the "automatic motor turning feature" after setting your speed CVs:
The locomotive will start out at full throttle but will quickly come to a stop. I did this for a Loksound Select decoder in a brass H20-44 and was getting well under 1sMPH at speed step 001.
Tom
This CV 54 thing is very interesting. where can i find out more about it?
Do you do this with all your locomotives?
Gary,
No, I only performed the test on the H20-44 switcher because I was getting a jerky response after installing the Select decoder. (I had previoius installed and tested a TCS motor-only decoder in it and it ran as smooth as glass.) The majority of the time the Loksound decoders are just great right out of the box - i.e. <1sMPH on speed step 001. Other than setting the address, turning off dual-mode, and apply Rule 17 lighting, I hardly ever have to tweak any other CVs.
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
The CV54 Auto Tune sets the four different BEMF settings automatically based on feedback from the motor itself. Unless you REALLY understand how the BEMF variables are calculated, it can be a real crap-shoot to try and set them perfectly manually.
Mark.
¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ
See the Select (Direct, 6Aux and Micro) manual pg. 22-23. Download it below. Or the V4.0 manual if that's what you have. The info is similar. It explains how non-typical motors might want to try different (than default) settings for BEMF CVs 52-55 for can motor w/o flywheel or coreless motor, as well as the CV54 procedure for auto-tuning, as well as individual CV tweaking.
http://www.esu.eu/en/downloads/instruction-manuals/digital-decoders/
I have only done the CV54 procedure for a few locos, but I will get around to doing it on each eventually (perhaps). I did find the unique NWSL Stanton drive needed help. When speed was dialed down, it lurched to a multi-point stop, totally unsatisfactory. The CV54 procedure helped it immensely, though still not perfect perhaps. The individual CV tweaking (trial and error) I don't recall helping much additionally. Again, most locos run at slow speeds just fine out of the box. I only had to implement the CV54 procedure about 5% of the time.
Paul
Modeling HO with a transition era UP bent
Gary
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Mark,
You say the ESU only allows 26 entries in the speed table, as opposed to everyone else's 28. And yet, for my ESU decoder, I was able to change the first speed step (CV 67) from the delivered value of 5 (apparently a special case--see below) down to 1.
You appear to be saying I can't do that, but must use CV2. And yet my CV readout says it happened.
So I admit I am not following you.
There is currently momentum in the throttle right now. And it feels to me to be about the right amount. In particular, I don't want to increase it.
It appears that the seller programmed this decoder themselves. Which gives me some interesting variables to deal with. So far, nothing really irritating, though. But that might explain why CV 67 was delivered at 5 instead of the linear value of 9.
Ed (an admitted beginner)
ESU (Loksound) handles speed curves differently than all other manufacturers. With all other manufacturer's speed table, you have 28 points of reference you can set. With ESU you only have 26 that are settable in the speed table itself. You don't set point 1 or point 28 in the speed table - you set those using CV2 and CV5.
Why ? .... well, it's rather eloquent when you think about it ....
You spend a good deal of time creating the perfect speed curve. Now you want to lower the top speed. Normally, you would have to re-set every point on the speed curve to match new lower value on step 28. With ESU, you just lower the top speed using CV5. The internal logic automatically adjusts all the other steps to retain your original curve, just a bit shallower to match the new lower setting of the top speed set using CV5.
Same goes for your start speed. You don't set step 1 in the speed curve, you set it with CV2. If you are still getting a jump start, add some momentum in addition to the BEMF.
rrinker How are you actually controlling it, with 28 steps or 128? Try 128, much finer control. --Randy
How are you actually controlling it, with 28 steps or 128? Try 128, much finer control.
128.
First of all, one can enter a value of 0 into speed step 1 (of 28). But the program changes it to 1. So, my change of speed step 1 from value 1 to value 0 never really happened.
I just timed the loco. With speed step 1 (of 28) set at zero, it travels at 1.3 mph at speed step 1 (of 128).
The first several speed step values were 1, 5, 8, 11.
Then I wondered what would happen if I changed those values to 1, 2, 3, 4. I thought maybe with a shallower curve slope, it would lessen the absolute beginning speed. I notice no change: 1.3 mph.
So, I'll change the 1, 2, 3, 4 values back to the previous values.
It seems clear the motor is willing to spin slower, because it is REAL happy just ticking over right now. At 1.3 MPH.
Thanks guys,
I'll pursue this later.
tstageDo you really want to mess with speed tables? You might try setting CVs 2, 5, and 6 first to see if you can achieve what you want. It's actually quicker than speed tables and I've gotten very good results.
I totally agree, I have never needed to use the umpteen settings speed table approach. Setting CV2 (slow) at speed step 1, CV5 (max) at speed step 126 first is easy to set the range. Then, at speed step 63, I set CV6 (mid) to get the slower speeds slope and the faster speeds slope. That works great for speed matching locos.
tstageYou might also try running the "automatic motor turning feature" after setting your speed CVs: Place locomotive on minimum 3' piece of track Change CV54 to a value of "3" Press F1 The locomotive will start out at full throttle but will quickly come to a stop.
Yes to this also, which fine tunes BEMF variables, except (per the Select manual, pg. 23) set CV54 to zero (not 3) before hitting F1. That will make the loco self-adjust CVs 52-55. Best to do this first (on a warmed up, broken in loco) and then set CV2 for smooth movement at speed step 1. Then set CV5, then CV6, per above.
Loksounds usually run nicely, including at slow speed, out of the box before doing the CV54 trick. I have converted a number of P2K with Selects and usually do not even bother with the CV54 trick. On a Doodlebug with a Stanton drive mod it was very necessary as the drive lurched significantly. If the CV54 process does not give decent speed step 1 operation via CV2 adjustment, maybe there's a drive issue. Note that CV2 is not necessarily set to 1 for smooth operation at speed step 1; default is typically 3, but see what gives smoothest low speed operation. It depends on the loco.
If speed matching two locos, one likely will run smoothly a bit slower than its mate when both are adjusted at speed step 1 with CV2. In that case, the slower one's CV2 can optimally be increased slightly to run at the faster speed of its mate. That's a fine point, it might not be necessary, but that's what I do.
Let me know how it goes for you, Ed.
BTW, the value I originally posted above for CV54 for the automatic motor turning feature should actually be "0" rather than "3". I made the correction in my original post.
Tom,
I've done the speed tables a couple of times so far. Yeah, it takes a few minutes; but I kinda like it.
Thanks for the info/suggestion. I'll be thinkin' on that.
Meanwhile, as an experiment, I changed the CV in speed step 1 from 1 to 0. I'd been told that couldn't work, but it does seem to have improved it. Not to perfection, though.
I do hafta say that the loco is going pretty slow, though. But what bugs me is that it still sorta just STARTS. Irritating.
Thanks again for the suggestion. It's gettin' late, but I'll try it tomorrow (-ish).
I'm just programming a P2K GP9. It's got a Loksound decoder. I'm using the speed tables. There's 28 entries to be used. I've got speed step one programmed to 1. My irritation is that the loco jumps up from zero to a certain speed. And my further irritation is that I don't see how I can get the starting speed step below 1.
Am I missing something here????