|
Post by BAZman on Jun 11, 2021 16:35:38 GMT -5
If they run well in DC then these would be my thoughts of why you have poor running with DCC:
1) BEMF (back electromotive force) a.k.a. Load compensation. TCS Is notorious for this effect. Make sure when you programmed it that you turn off the bit for BEMF
2) CV 3&4 (acceleration, deceleration) has a non-0 value, meaning it is trying to SLOWLY start up your locomotive and slowly slow it down.
3) PC Bd contacts to the posts. Probably just loose enough to vibrate off the contact to post. And/or the PC boards to the motor tabs.
4) gobs of grease in the trucks and/or worm gear.
The DCC board capability needs to have a meeting-of-the-minds to make this easier than scraping solder off the contacts, until they fit (properly)
You’ll like still have so issues with the Digitrax but, it’s a new data point.
|
|
|
Post by husafreak on Jun 11, 2021 23:41:12 GMT -5
Thanks Jeff, we'll see if the Digitrax boards do a better job in these locos. I've got a sharp eXacto blade ready! But to your suggestions. !. Done per your instructions. 2. Can't remember ever changing these. But I gather that 0 is ideal. Next rainy day I'll check. I never noticed the effect, the trains followed my throttle inputs. 3. I worked on this quite a bit, per your instructions, I do know they were a tight fit. 4. No.
|
|
|
Post by husafreak on Jul 9, 2021 22:20:42 GMT -5
I installed the DZ123Z0 boards in my two AZL ES44AC locos today and they both ran great! I should say that both boards required a bit of solder carving with an X-Acto before they would slip into place and even with the motor contact tabs at 90 degrees to the board they still needed to be pulled down a fraction to get a good contact with the motor tabs, so not exactly "plug and play". but once they were moving they kept moving! I did no other programming besides assigning addresses and making a consist of them. This ease of operation was also my experience with my Digitrax equipped MTL loco. To quote the great RRay "they just work". Smooth and consistent slow or fast, lights on steady. End of story for these locos Jeff thanks again for trying to help me with this. I know you are right, the TCS AZL4 boards have got to work. I will try again with them! Maybe they were just incompatible with these loco's. Bad Ju Ju or something... I'll find another loco that they can be used with and I'll reset them to factory specs and try again with your advice, they must work sometimes right?
|
|
|
Post by BAZman on Jul 10, 2021 1:07:44 GMT -5
Good to hear. Whether the Installation, Programming or just the characteristics, its difficult to quickly resolve. Itemizing:
Board mechanical connections: 1) to the chassis post for electrical pickup, with the varied removal of solder to the motor contacts 2) interference of the board components: a) preventing motor contacts b) inability to mechanical contact to the notches on the back posts (so the board pops out)
As for programming, I mentioned: CV3 & CV4 (Acceleration and Momentum) to Zero as with power issues (track, wheels, pickups and/or the boar-to-frame contacts) BEMF variables (just 1 CV value for TCS, up to 3 for Digitrax) but TCS has been the biggest by far as its algorithm doesn't work for Z. I have been pleased with the lack of BEMF issues on Digitrax DZ123Z0's
[ I couldn't get the Bullet List for work right and I could not find Indent either so the post format will be sloppy]
|
|
|
Post by bogtudor on Dec 27, 2021 16:20:14 GMT -5
I had the same experience with an TCS AZL4 installed in a SD70Ace, it shuttered no matter what I did. However, it did run perfect with the DC board and with a Digitrax decoder. Also, I have a TCS AZL4 installed in a 2nd gen SD70M and it runs perfectly, go figure. I noticed that the power pickup system of the 1st and 2nd Gen SD70M is more refined even with the old motor tabs design. Also the pickup strips metal seems different too, the Ajin made ones are more elastic and have a different color. My guess is that TCS needs perfect electrical pickup while Digitrax is more tolerant oxidation dirt etc. Now I put No-Ox on all my locomotive pickup strips near the contact points to prevent oxidation.
|
|
|
Post by husafreak on Jan 11, 2022 13:28:06 GMT -5
I started a new thread closely related to this one: Different versions of AZL GP7 and DCC conversion. In most ways that was a mistake because if I had reread this thread from the beginning I would not have needed to post at all! Apologies to the "old guard" here for making them repeat themselves. I was clearly told that my AZL GP7's were older versions and would require TCS Z2 boards but, well, that was two years ago and I forgot...
As for what this thread morphed into I can repeat that since installing DZ123Z0 boards in my AZL ES44AC's I have had no running issues whatsoever. I would like to enable the "momentum" feature on those DZ123Z0 equipped locos though as I do like that feature a lot, especially when changing direction. As always any advice would be appreciated.
Good news is that the TCS AZL4 decoders found new homes in my recently purchased AZL GP-38-2 locos and so far they are working well there, maybe not as rock solid as the Digitrax locos, but I can live with the occaisional flicker at slow speeds over turnouts, hey, they keep moving!
I ordered a couple of TCS Z2 wired decoders for my "old version" GP7's so hopefully soon I can put this old thread to bed! Tons of great info here though, maybe useful to many others in the future.
|
|
|
Post by husafreak on Jan 28, 2022 23:25:25 GMT -5
I installed the TCS Z2 decoders in my two "old" AZL GP7's today, the ones with the Falhauber motors. This was my first experience with wiring a decoder. From various queries on the forums and the great advice given I was all set to go with my Weller soldering station and fine point tip, 015" solder, Optivisors, and Dremel with dentist size bit. I won't say how long the first one took but the second was installed, programmed, and running on the main in one hour. They both run very well but one is slower than the other, it moves right along on step one versus the other which crawls along at step one. Probably need to compare the CV's on those two and see why, something may have slipped by me. I learned a lot. The first installation had me scratching my head, on the program track the loco was running very strangely, and in reverse it wouldn't stop running! WTheck? At first I convinced myself it was a programming issue. I factory reset the board and programmed or checked every CV I understood without luck. When that proved fruitless a careful examination of the physical work showed a slight melt on top of the white wire where it crossed the grey one. I pushed them a bit and saw they were stuck together. Knowing how easily the wire coating had melted and pulled back during the tinning process I thought I might have over cooked it. That made me wonder if the two wires may have contacted where I got a little too much heat when soldering the white one last. I redid that connection with a piece of Caption tape between the wires and presto! A perfectly functioning loco For the next installation I lowered the iron temperature to 600 degrees from 650 (.015" lead and thanks again for "soldering school" Jeff) and made sure to work quickly when soldering this crazy thin stuff. The installation was a snap the second time around. But on the program track the fun began! It ran great on short address 3 but I could not program a long address. I tried over and over, it just would not take a long address. During this process It started bugging me that the short moves that locos do during programming were exaggerated with this loco. Instead of moving a few mm it would go a couple of inches. I remembered that on my first installation I had changed CV61=0 before setting its long address so I went ahead and changed CV61 while the loco was on short address number 3. Lo and behold the movements during the next long address programming cycle were what I expect to see and the loco happily went to work with its long address! Jeff, again, you are the man! I knew it was important to set CV61=0 but in this case it was necessary to do that before the decoder could be programmed at all. I think! So this business of wiring tiny decoders is now known. Dang, this is some small stuff physically, and programming can be a riddle. But it was fun and the TCS website "how to" photos and instructions were very good. And good for AZL to provide (at least on these locos they did) a DC board with pads ready to be used in a DCC installation! That is really great. Time to tackle my Mikado! The tiny bit of running I did tonight showed the TCS Z2 to work very well in these locos on the main track. They appear to be running very smoothly. As always thanks everyone for your help.
|
|
|
Post by BAZman on Jan 29, 2022 2:05:42 GMT -5
hmmm. Thanks for the props. There should be NO reason that you can't program the Long address. On NCE, you missed a step. After entering the Long Address (AND pressing 1 to Activate this (long) Address), the next prompt is: Set CFG? (1 for Yes)
So, this would be start to finish:
Esc, 4 (jump to Program Track), Enter. Choose 1 (Standard programming), Enter Enter (manufacture) Enter (version) Enter: should say "Active Address" on top Setup Address? 1 (yes) Short Address? Enter (skips) Activate this Address? Enter (no) Long Address? enter your road number Activate this Address? 1=Yes
Make sure you do this Prompt: Set Config? 1=Yes
Enter (Default "normal Direction" i.e. Forward) Enter (Default "Speed Step=28" 1 =DC mode (so you can test with a 9 volt battery) Enter (Speed Table=Std) 1 =Address? LONG
That last one is like, "OK, Done"
|
|
|
Post by bogtudor on Jan 29, 2022 6:44:59 GMT -5
|
|
|
Post by husafreak on Jan 29, 2022 13:08:22 GMT -5
Thanks guys. 1. I’ve never had a problem programming a long address before. And I did get out my NCE manual and follow the steps exactly. After entering the long address and pressing 1 to activate it I was getting the message “ CAN NOT READ CV” . Many times. 2. Then all I did was change CV61=0. 3. Then the loco accepted the long address. Yeah I accept that there might have been something else going on or error on my part but at least through my fumbling I got it going.
|
|
|
Post by husafreak on Jan 29, 2022 13:10:02 GMT -5
Thanks for reminding me to enable DC mode. I did not do that but I have on other locos so I want them all the same
|
|
|
Post by husafreak on Jan 29, 2022 13:16:44 GMT -5
Apologies for multiple posts but note that these locos have 3 digit road numbers so I am entering 0113 or 0122 and then seeing #113 or #122 on my Powercab. Fine with me I don’t see why that would affect anything.
|
|
|
Post by ptitrainrouge on Jan 29, 2022 17:05:56 GMT -5
I have the same pb with one of my loco
tomorow, I will try to put the cv61 to 0 and see
thanks for the information
|
|
|
Post by husafreak on Jan 29, 2022 18:34:10 GMT -5
Look at BAZman posts he explains why CV61 does not work well with our small Z scale locos.
|
|
|
Post by BAZman on Jan 30, 2022 1:46:27 GMT -5
Lack of details = game of darts There is a difference in “long address” in DCC. There is Short Address 1 thru 127 (the original way back when [still can be used . . . read on] and Long Address 0001-9999 Generally, no one uses Short addresses. With the NCE system, you could still use the Short Address for the 113 & 122 but they prefer not to use the ‘1’ when Selecting Loco and only the 13 or 22. However, if you want to use the Long Address, you will need to put a ‘0’ before (0113) which is the way DCC interfaces, due to imply an 4-digit Address structure. NCE display’s the * as you found.
I have no idea why CV61 matters. After you do the Config (and Motor), then bailing out, it should run the loco (just Increase the Speed Step). It works when I did it.
While I’m typing, if you are going to Consist, use the Old style, so you don’t change CV19 and run from a 127 Consist number. The Old Consist uses the ‘Command Station’ (in the PowerCab), to pair up the locos at their own road number. That way, when there is a problem with a loco, just Del and Add back or another. Plus easy to Program on the Main (POM) to change say CV 6 or 5 to help speed match (while they’re running separate)
|
|