Happy New Year to everyone:
Two of my Kato SD90MACs have quit. All I get is one LED light on, indicating reverse. They each have a basic Digitrax DH126 DCC decoder. I've checked the decoders on others locomotives and with the Broadway Decoder tester and the decoders are fine. I've even tried three different DH decoders..same thing with the SD90MACs - nothing happens from the throttle.
I switched two circuit boards from other units that work fine, thinking maybe that was it...nope (and CBs no longer available at Kato). I also made sure they were connected to the circuit board as per the advisory years ago.
Checked CVs in JMRI and JMRI read the chips from the two SD90s fine! Also tried writing back and the lights flashed and could hear the clicking of the decoders.
Would this not operating be the motor if the decoders work fine? The units get power and shorts out the block so power is getting through. Or something else? If it is the motor(s), could someone suggest a good DCC motor replacement. I am just beside myself why two would suddenly both quit.
Thanks for any help!
Robbie
Seems to me that if you had the units on the track, with the "body" off so you can easily access the electronics, that you could put a voltmeter on the board outputs to the motor and see if there is a DC voltage.
Of course, you have to tell the decoder via your throttle to do that.
If you get a voltage, then the problem is likely downstream: the motor and/or the wires and connections getting there.
If you don't, (and you should), then the problem is the decoder, or before.
You could also remove the motor and zap it with 12V DC from ye olde power pack--see if it spins up.
Ed
Unplug the decoder and re-install the DC dummy plug and see if the engine runs on DC. If not, diagnose why. I it won't run on DC, it isn't going to work with a decoder in it.
Mark.
¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ
If you select the loco and hit the direction switch, do the lights change? If not, I'm going to have to say the commands are never even reaching the decoder. Are you sure you are using the correct address? If you can read and write the decoders, try resetting them by setting CV8 to 8. The loco should work on address 3.
Were they ever used in an advanced consist? Make sure CV19 is 0.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Hi:
THANKS to everyone for the suggestions and ideas - tried the motor free of the frame and it works fine for both locos. Also checked leads to the trucks and everything looks connected after put it back in...same thing happened. One light, no power.
In another post, only the rear (reverse) light lights. Your idea about the address got to me. I checked it using the Broadway Decoder reader and also JMRI again and all say 118 and 123 from the decoders. Checked CVs and all matched. CV 17/18 showed the correct long address (118 & 123) - I did try 03 and nothing.
BUT above gave me an idea hadn't thought off...what if I use the file of another SD90MAC that works without issue? So I copied the file of 9120 to 118 using the file I had made...BOTH LOCOs WORK now!!!!
Now, I tried to change the address (long) on the programming track...it clicked, lights lite - put back on mainline...dead!!!!!!!!! Back to copying the file onto the decoder using 9120 file and sure enough 118 & 123 work again!!! SO WEIRD!
NOW I can't change the address via the programming track. Haven't tried in JMRI yet but might have to do that too? Would the address section of the decoder be corrupt somehow so can't change? Can an address be locked maybe??
Thanks to all for the help and your time! So much appreciated!
Happy New Year!
My worth, so you took out the motor, and the bare motor runs fine. OK, I would still do what Mark R. above suggested, remove the decoder from the board, install the jumper plugs back in, and see how the locos run in DC.
Unless I'm completely off base as to what type of decoder you have, the Katos I have use a simple plug in, so it's quick and easy to unplug the decoder and install the jumper plugs.
Mike.
My You Tube
What DCC system are you using and how are you setting 118 and 123? 118 and 123 are only operable on Digitrax as short addresses, CV29 less than 32. With Lenz, they MUST be long addresses, CV29 greater than 32. With NCE< they can be either, but the way you select them on the throttle differs.
It's the one spot the NMRA really dropped the ball on setting the standard, allowing different systems to so things differently and in incompatible ways.
Obviously both locos are Ok. I suspect CV29 is wrong. I also guess you are using an NCE system. Anyhow to program a long address on these locos here are the values:
Loco 118 CV17=192 CV18=118 CV29=34 analog off or 38 analog on.
Loco 123 CV17=192 CV18=123 CV29=34 or 38
Also make sure CV19=0 on both.
Personnally I never use a long address for value below 128 with my NCE Power Cab.
Jack W.
jalajoie Personnally I never use a long address for value below 128 with my NCE Power Cab.
I programmed good ole GN 1 with my NCE to a logical* long address of 0001. Worked great. Went to a Free-mo setup which uses Digitrax. Nope.
Did research.
Used my same NCE to program ole GN 1 to 01. Worked great. Here and at the next setup.
Now, the sticky part is that Digitrax uses the short addresses for consisting, typically backwards from 127. 1 is a LONG way from 127. But if you have a loco with a number up above 100, or so; you need to do further research. SP&S, by the way, had numbers up to 101, and then skipped to 150. Getting REALLY STICKY, RF&P went right up to 127. And I have RF&P 126. Yikes!!!!!
*logical in that it seemed thus to program ALL locos with a long address. For simplicity and straightforwardness. Yup, it did.
Hi everyone - thanks for the super help! So much appreciated.
So I tried each of the different suggestions each of you offered. I've been using Digitrax from almost the very beginning and all locos with Digitrax decoders except I tried a TS1 decoder one time...mistake. Only one I ever smoked.
The last suggestion to get this power moving was the short address. Have just lifted the locos off the programming track with the five-finger hook and the short address of 18 and 23 WORK! I am totally flabbergasted...and somewhat embarrassed. I never gave it a thought to try that...THANK YOU ALL for your help.
I had NO idea about the consisting issues with short addresses. For "fun" I'll have to try that and see what next I can screw up.
I made a slide show of model railway and put on my Telegraph site via my GO site - I need to add other images but if anyone is interested it is at:
http://www.trainweb.org/telegraphgene/GO_MR/go_modelrwy.htm
Again, thank you to everyone. Really appreciated! have a Happy New Year everyone - Robbie
7j43k jalajoie Personnally I never use a long address for value below 128 with my NCE Power Cab. I programmed good ole GN 1 with my NCE to a logical* long address of 0001. Worked great. Went to a Free-mo setup which uses Digitrax. Nope. Did research. Used my same NCE to program ole GN 1 to 01. Worked great. Here and at the next setup. Now, the sticky part is that Digitrax uses the short addresses for consisting, typically backwards from 127. 1 is a LONG way from 127. But if you have a loco with a number up above 100, or so; you need to do further research. SP&S, by the way, had numbers up to 101, and then skipped to 150. Getting REALLY STICKY, RF&P went right up to 127. And I have RF&P 126. Yikes!!!!! Ed *logical in that it seemed thus to program ALL locos with a long address. For simplicity and straightforwardness. Yup, it did.
Other way around, NCE is the one that uses short address 1-127 for consist addresses. Digitrax doesn't necessarily use ANY address but the loco's own for consists.
Since CV19 only takes 1-127 (129-255 replicate those but mean run the other direction), you don't have much choice, they are the same address as the short address.