I seem to have narrowed my problem with locos that program OK, run with EasyDCC throttles, but not with WiThrottle.

Every one so far has the same brand and model decoder: an early Lenz LE103 unit. They were my choice when I got started with DCC, reasonable size, price, and current rating. Most of my locos have Lenz LE103 decoders, and NONE tested so far works with WiThrottle! DecoderPro identifies them, programs all the CVs, they run with EasyDCC throttles set to those addresses, but don't run with WiThrottle. Given I have ten or more locos with LE103 decoders, I sure wish WiThrottle could control them!

The two locos that are working are using other decoders, one an inexpensive NCE bought a month ago, the other an older Zimo, probably an MX-61N. DecoderPro identified the NCE and its model correctly. It couldn't identify the Zimo, even by brand, yet could program and control it.

Has anyone else had problems using Lenz LE103s or other older decoders with JMRI? Does anyone know why this should be happening, or better still, have a fix?

Don Weigt
Connecticut, USA

开云体育

Don

I will test tomorrow - I removed these decoders form locos years ago - some are now in cabooses (cabeese) to control the lights.

Gerry

I seem to have narrowed my problem with locos that program OK, run with EasyDCC throttles, but not with WiThrottle.

Every one so far has the same brand and model decoder: an early Lenz LE103 unit. They were my choice when I got started with DCC, reasonable size, price, and current rating. Most of my locos have Lenz LE103 decoders, and NONE tested so far works with WiThrottle! DecoderPro identifies them, programs all the CVs, they run with EasyDCC throttles set to those addresses, but don't run with WiThrottle. Given I have ten or more locos with LE103 decoders, I sure wish WiThrottle could control them!

The two locos that are working are using other decoders, one an inexpensive NCE bought a month ago, the other an older Zimo, probably an MX-61N. DecoderPro identified the NCE and its model correctly. It couldn't identify the Zimo, even by brand, yet could program and control it.

Has anyone else had problems using Lenz LE103s or other older decoders with JMRI? Does anyone know why this should be happening, or better still, have a fix?

Don Weigt
Connecticut, USA

-- 
Gerry Hopkins MMR #177 FNMRA
Great Northern Downunder




NMRA Australasian Region
Contest & AP Chairman
Web Administrator

Mike, I'm having this problem with an EasyDCC system. I'd be interested in your idea when you have access to something with a keyboard.

Gerry, I look forward to your test results! I've thought about using the Lenz LE103s for lighting controls or my turntable. Of course, the turntable won't work if a loco won't. But, I've not even been able to control the lights on the four locos that have them. So far, I have control of lights and motor, or neither.

It's worth mentioning that when I bought these decoders, one consideration was that the complied with the NMRA DCC Standard. So, I expected them to remain compatible with newer DCC equipment.

Don Weigt
Connecticut, USA

My problem just got bigger. I put a new Soundtraxx Econami Steam xxxPNEM sound decoder in one of my locos. DecoderPro didn't find the brand or model in CV 8 and 7. It reported both contained zeros. Yet, it identified a Lenz LE103 again...

When I set the decoder type, I was able to program it. I can run it with my EasyDCC plug in throttles, but not with WiThrottle through my RPi running Steve Todd's image, a serial connection, and my EasyDCC Command Station.

I have a SoundTraxx programming booster, but haven't tried it. I read it's not needed for the newer Econami decoders.

So, my problem isn't just with one old type decoder. It includes at least one modern sound decoder. There's a clue there, somewhere. But, for the moment, it's discouraging.

Don Weigt
Connecticut, USA

Don

Did some testing yesterday with Lenz 103, Lenz 103 XF, NCE 102 and System One 102EU.

First, using a decoder tester (for ease of connection) and Sprog 3Lenz 103. DP identified the decoder correctly. I did a full reset and entered a short address of 66. I then used the throttle in DP to test the decoder it did run - but not very smooth. I then set it to 14 speed steps and ran it again with DP - it ran.

Next step, Using my Easydcc CS2 (22 year old command station with V6.29 firmware - I ran the decoder through the same steps with the tethered throttle. Same results.

Step three, set the computer to Panel Pro and turned on the WiThrottle server and used my phone. It did the same as above with the tethered throttle.

Step four, as all the above with output of the command station set to 14 speed steps ran OK. Set to 28 ss and it ran OK. Set to 128 ss and it ran jerky. The Default on the command station is 128 for all modern decoders. By modern I mean those with "silent drive".

I did the same with the other "old" decoders and the results were the same.

Then the big test - I did all the same tests but this time with an address of 4567. The decoders responded the same with an exception - The phone would not run the decoders when the command station was set to 128 speed steps. I had the same results with another command station.

At the age of 74 there are a lot of things I have forgotten about early decoders. This said, I believe the "problem" has to do with speed step setting along the path - similar to the headlight blinking in 14 speed steps.

The System One and the NCE behaved the same - BUT - the Le103XF worked through all combinations.

Today I will try again with the Raspberry Pi 3B Sprog.

Gerry


--
Gerry Hopkins MMR #177 FNMRA
Great Northern Downunder




NMRA Australasian Region
Contest & AP Chairman
Web Administrator

Mark,

Thanks for the suggestions!

WiThrottle seems to not be my problem controlling Lenz LE103 decoder equipped locos. I opened a JMRI throttle on my laptop VPNed to the RPi, I think this is what you meant by "Virtual Throttle." It was able to control the same locos as WiThrottle and EngineDriver. The locos EngineDriver could not control, also couldn't be controlled by the JMRI throttle on my laptop.

Not reading CV7 and CV8's values on my old Zimo and new SoundTraxx Econami decoders is an EasyDCC problem. When I used the Command Station's front panel controls to read them, EasyDCC reported zero values on its front panel display. EasyDCC is unable to read those CVs correctly. DecoderPro is reporting what the EasyDCC is finding. I'll scope the programming track waveforms and see if any problems are visible doing that.

I could control the SoundTraxx Econami equipped loco with an EasyDCC plug in throttle, and also from the JMRI "virtual" throttle, and also using Engine Driver connected to WiThrottle. I thought yesterday that it didn't work... Hmm.

So, the remaining problems seem to be: my EasyDCC can't read CV7 and CV8 on Zimo and Econami deciders (and perhaps more CVs), and my RPi/JMRI system can't control Lenz LE103 decoders through my EasyDCC.

Gerry,

If the number of speed steps set makes a difference, you should know I have been setting all my decoders to 28 speed steps. Thank you for all the testing you're doing! I really appreciate it. It's almost as good as having known good units here I could swap for my stuff until I find one that fixes the problem. It's quite amazing, when the distance between us is considered....

Don Weigt
Connecticut, USA

开云体育

Don

On the command station you can select 14, 28, 128 these are the speed steps sent to the decoder - not what is set in the decoder. Using JMRI you get the choice of 14 or 28/128 for the decoders but no selection for the output. I had a Digitrax DN120 in one of my narrow gauge locos (#10) If the command station was set to the default of 128 the decoder would respond with speed only - no momentum and no speed table. Earlier this year it was upgraded to a TCS M1. On the Command station a single loco can be set for a specific speed count so I was able to get round the running of the 200mph narrow gauge loco - I had to remember to set the speed steps after a reset of the panel.

Something I have not asked - do you run a program booster from the Easydcc unit such as Power Pax or PTB100? Also check your panel - on the back near the program terminals there is a resistor - R7 - it is 10 ohms. If it is black or discoloured replace it.

Now for something unexpected - when testing the various systems - Using the WiThrottle on my Samsung, through JMRI then through EasyDCC panel I found I could use up to F18 - where all the Easy DCC throttles and panel throttles can only go up to F12.

Gerry

Gerry,

If the number of speed steps set makes a difference, you should know I have been setting all my decoders to 28 speed steps. Thank you for all the testing you're doing! I really appreciate it. It's almost as good as having known good units here I could swap for my stuff until I find one that fixes the problem. It's quite amazing, when the distance between us is considered....

Don Weigt
Connecticut, USA

-- 
Gerry Hopkins MMR #177 FNMRA
Great Northern Downunder




NMRA Australasian Region
Contest & AP Chairman
Web Administrator

Going from memory, Lenz LE103 decoders supported a maximum of 28 speed steps. ?The later LE103XF were an upgrade that brought in (amongst other things) 128 speed steps.

Steve

On 3 Sep 2019, at 1:59 PM, Don Weigt <dweigt47@...> wrote:

Looking at the programming track waveforms, they look good. Quite square (fast rise and fall times) and nearly 12V for the positive parts and near 1V for the negative, with no load. Slightly higher low voltages and lower high voltages with a Lenz LE103, and slightly less again with the Econami. About 11V and 1.5V with the Econami. (It's a little hard to read closely on my old analog scope.) The difference between the two rails seems to be nearly 10V. one going higher when the other goes lower.

DCC waveforms should be symmetrical bipolar with always around 12-14V between positive and negative.

It's usual for one rail to go to ground at a time and the other to +12-14V at the same time So for this reason you shouldn't connect a normal CRO probe with ground clip across both wires, unless your CRO is fully floating. A differential input probe pair is preferred, or you can connect the ground clips to DCC system ground and use two channels (that's what I do with my old analog CRO). I also use a DSO adaptor on my iPad, which means my inputs are fully floating.

For typical waveforms, see NMRA S9.1 (on the NMRA website).

Dave in Australia