Topics

2010 Atlas Gold TrainMaster randomly stops and restarts


Dave Horn
 

I have a 2010 #7883 2010 Atlas Master Series Gold EL TrainMaster cab #1860 that I like very much except for one annoying thing that it does - it randomly stops, sounds quit, lights go out, and then it just that quickly starts itself up again and gets back underway with throttle sounds ramping up, speed increasing, all other sounds going, lights back on as if just getting underway up to the speed it was at when it stopped.  It does this in in RTC mode or STC although a bit less frequently in STC than RTC.  It does it on straights or curves, running clockwise or counter clockwise.  It does this after 8 minutes, 2 minutes, or 20 minutes of running.  There's no rhyme of reason to the occurrence of this annoying "stall".  CV reset and reed-switch reset do not stop it from happening again.  I've inspected wheel gauging, cleaned the tracks, cleaned the wheels, inspected all wires to and from and connections on the board and trucks, inspected all plugs on the board, ran it with the speakers disconnected in case one of them was defective and overloading the decoder, added MG Chemicals conductive grease #846 to the axle tips, ran it with the sell off to see if anything inside gets hot, and still it momentarily conks out and restarts itself unpredictably at will.  What do you suggest is wrong?  What do you recommend I do?  I'm running it with address 3 using a NCE PowerCab DCC system with which I run BLI, InterMountain, Athearn Genesis, and other DCC engines none of which have this problem.  Thank you.


litespan46
 




On Sep 11, 2020, at 8:10 PM, Loco loco via groups.io <davidghorn@...> wrote:

I have a 2010 #7883 2010 Atlas Master Series Gold EL TrainMaster cab #1860 that I like very much except for one annoying thing that it does - it randomly stops, sounds quit, lights go out, and then it just that quickly starts itself up again and gets back underway with throttle sounds ramping up, speed increasing, all other sounds going, lights back on as if just getting underway up to the speed it was at when it stopped.  It does this in in RTC mode or STC although a bit less frequently in STC than RTC.  It does it on straights or curves, running clockwise or counter clockwise.  It does this after 8 minutes, 2 minutes, or 20 minutes of running.  There's no rhyme of reason to the occurrence of this annoying "stall".  CV reset and reed-switch reset do not stop it from happening again.  I've inspected wheel gauging, cleaned the tracks, cleaned the wheels, inspected all wires to and from and connections on the board and trucks, inspected all plugs on the board, ran it with the speakers disconnected in case one of them was defective and overloading the decoder, added MG Chemicals conductive grease #846 to the axle tips, ran it with the sell off to see if anything inside gets hot, and still it momentarily conks out and restarts itself unpredictably at will.  What do you suggest is wrong?  What do you recommend I do?  I'm running it with address 3 using a NCE PowerCab DCC system with which I run BLI, InterMountain, Athearn Genesis, and other DCC engines none of which have this problem.  Thank you.


Jay Beckham
 

I also have two Atlas Gold Series GP 9 that stops and starts on their own.  Best I and others can tell is the NS plating has worn off the wheels and that is causing loss of contract.  I have ordered new wheels from NWSL but it takes them months to get from China.  So far it has been a month.  Also check the pickups as mine were doing a poor job of pickup.  Also tried conducting oil which helped some.

 

I am told that the base metal of the wheels when in contract with NS rail causes corrosion and that is the problem.  If the new wheels don’t fix the two engines I will replace the QSI decoders with another brand.

 

You could also try resetting the decoder to factory with the wane or using NCE.  Can’t recall the exact procedure but Google resetting Atlas QSI decoder and you should find the information.

 

 

Jay Beckham

 

My Blog https://jaysoscalelayout.blogspot.com/

 

Join us at JayOScaleLayout+subscribe@groups.io

 

 

 

From: QSIndustries@groups.io <QSIndustries@groups.io> On Behalf Of litespan46
Sent: Friday, September 11, 2020 8:30 PM
To: QSIndustries@groups.io
Subject: Re: [QSIndustries] 2010 Atlas Gold TrainMaster randomly stops and restarts

 

 



On Sep 11, 2020, at 8:10 PM, Loco loco via groups.io <davidghorn@...> wrote:

I have a 2010 #7883 2010 Atlas Master Series Gold EL TrainMaster cab #1860 that I like very much except for one annoying thing that it does - it randomly stops, sounds quit, lights go out, and then it just that quickly starts itself up again and gets back underway with throttle sounds ramping up, speed increasing, all other sounds going, lights back on as if just getting underway up to the speed it was at when it stopped.  It does this in in RTC mode or STC although a bit less frequently in STC than RTC.  It does it on straights or curves, running clockwise or counter clockwise.  It does this after 8 minutes, 2 minutes, or 20 minutes of running.  There's no rhyme of reason to the occurrence of this annoying "stall".  CV reset and reed-switch reset do not stop it from happening again.  I've inspected wheel gauging, cleaned the tracks, cleaned the wheels, inspected all wires to and from and connections on the board and trucks, inspected all plugs on the board, ran it with the speakers disconnected in case one of them was defective and overloading the decoder, added MG Chemicals conductive grease #846 to the axle tips, ran it with the sell off to see if anything inside gets hot, and still it momentarily conks out and restarts itself unpredictably at will.  What do you suggest is wrong?  What do you recommend I do?  I'm running it with address 3 using a NCE PowerCab DCC system with which I run BLI, InterMountain, Athearn Genesis, and other DCC engines none of which have this problem.  Thank you.


Donald Goveas
 

A small correction in the previous mail. The phrase 'decoder system' should read as 'DCC system'.
Donald

On Sat, Sep 12, 2020 at 10:29 AM DONALD GOVEAS <goveas.donald@...> wrote:
The best way of determining whether the decoder misbehaves or the pick up system defaults, is to mount the engine on EZ-Riders or any similar rollers and connect the decoder directly by wires to the decoder system. Run the loco with this setup for some time at various speeds and direction, interrupted with stops and starts.
I Hope this helps in diagnosing the root cause.
Best Regards
Donald


On Sat, Sep 12, 2020 at 5:40 AM Loco loco via groups.io <davidghorn=verizon.net@groups.io> wrote:
I have a 2010 #7883 2010 Atlas Master Series Gold EL TrainMaster cab #1860 that I like very much except for one annoying thing that it does - it randomly stops, sounds quit, lights go out, and then it just that quickly starts itself up again and gets back underway with throttle sounds ramping up, speed increasing, all other sounds going, lights back on as if just getting underway up to the speed it was at when it stopped.  It does this in in RTC mode or STC although a bit less frequently in STC than RTC.  It does it on straights or curves, running clockwise or counter clockwise.  It does this after 8 minutes, 2 minutes, or 20 minutes of running.  There's no rhyme of reason to the occurrence of this annoying "stall".  CV reset and reed-switch reset do not stop it from happening again.  I've inspected wheel gauging, cleaned the tracks, cleaned the wheels, inspected all wires to and from and connections on the board and trucks, inspected all plugs on the board, ran it with the speakers disconnected in case one of them was defective and overloading the decoder, added MG Chemicals conductive grease #846 to the axle tips, ran it with the sell off to see if anything inside gets hot, and still it momentarily conks out and restarts itself unpredictably at will.  What do you suggest is wrong?  What do you recommend I do?  I'm running it with address 3 using a NCE PowerCab DCC system with which I run BLI, InterMountain, Athearn Genesis, and other DCC engines none of which have this problem.  Thank you.


Dave Hastings
 

  I had an a BLI loco with QSI sound do pretty much the same thing, it got so bad I shelved it, joined this forum and Kelly Dorf said send it to him, he would see what he could do with it, a month later it showed up at my door fixed. It had something to do with the capacitors.
  Also had an Atlas with QSI sound that every time direction was reversed, it would short out the whole track, I use DCC Specialties circuit breakers, then it would start working again, sent it in to Kelly at the same time, and I think he said it was the capacitors also. All it cost me was the postage to send them in.
  Before you give up on it, I would contact Kelly, you can do a search on this group and find his email.
Dave Hastings


On Saturday, September 12, 2020, 01:07:33 AM EDT, Donald Goveas <goveas.donald@...> wrote:


A small correction in the previous mail. The phrase 'decoder system' should read as 'DCC system'.
Donald

On Sat, Sep 12, 2020 at 10:29 AM DONALD GOVEAS <goveas.donald@...> wrote:
The best way of determining whether the decoder misbehaves or the pick up system defaults, is to mount the engine on EZ-Riders or any similar rollers and connect the decoder directly by wires to the decoder system. Run the loco with this setup for some time at various speeds and direction, interrupted with stops and starts.
I Hope this helps in diagnosing the root cause.
Best Regards
Donald


On Sat, Sep 12, 2020 at 5:40 AM Loco loco via groups.io <davidghorn=verizon.net@groups.io> wrote:
I have a 2010 #7883 2010 Atlas Master Series Gold EL TrainMaster cab #1860 that I like very much except for one annoying thing that it does - it randomly stops, sounds quit, lights go out, and then it just that quickly starts itself up again and gets back underway with throttle sounds ramping up, speed increasing, all other sounds going, lights back on as if just getting underway up to the speed it was at when it stopped.  It does this in in RTC mode or STC although a bit less frequently in STC than RTC.  It does it on straights or curves, running clockwise or counter clockwise.  It does this after 8 minutes, 2 minutes, or 20 minutes of running.  There's no rhyme of reason to the occurrence of this annoying "stall".  CV reset and reed-switch reset do not stop it from happening again.  I've inspected wheel gauging, cleaned the tracks, cleaned the wheels, inspected all wires to and from and connections on the board and trucks, inspected all plugs on the board, ran it with the speakers disconnected in case one of them was defective and overloading the decoder, added MG Chemicals conductive grease #846 to the axle tips, ran it with the sell off to see if anything inside gets hot, and still it momentarily conks out and restarts itself unpredictably at will.  What do you suggest is wrong?  What do you recommend I do?  I'm running it with address 3 using a NCE PowerCab DCC system with which I run BLI, InterMountain, Athearn Genesis, and other DCC engines none of which have this problem.  Thank you.


Dave Horn
 
Edited

Dave Hastings -- Thanks.  I will research contact information for and reach out to Kelly Dorf.  Bad caps makes sense in terms of how it behaves.  

Donald G - thanks.  The wheels and rails are clean, 91% isopropyl clean, but maybe I'll try 3M green scratchy pad as well on both if oxidation needs to be removed, but no other engine has this problem so maybe not.  I like to avoid abrasives even those that don't score the metal b/c inevitably they produce dusty dirt of some kind.  

Jay - thanks,  your suggestion of the engine up on rollers with direct connection to the decoder makes sense too.  If its the decoder the issue would prevail plus too if its bad caps then it will fail as well, but at least it would rule out pick-up unless there's something unique about it traversing rails instead of running on rollers.

Dave Horn


Tom in Texas
 

Try setting CV11 to zero. Some DCC systems don’t refresh the commands as soon as others

Tom in Texas


Nick
 

Hello "Loco loco" (Please sign your posts, it makes it a little more friendly)

The one small piece of information you have not provided is, have you upgraded the "crippled" chip in the decoder ? The original OEM chips in the first series of Atlas and BLI locos were victims of the Mike's Train House fake lawsuit where he falsely claimed to have invented features used in the decoders. He naturally lost the suit but only after forcing years of legal battles. During those years, QSI was forced to disable certain features of the decoders like BEMF, speed tables and more I can't remember. Since manufacturers had already started production on those locos, QSI placed "temporary" and removable chips in the decoders. The plan was to provide free upgrades to owners after the suit was settled. Unfortunately, Mike used years of legal tricks and false info to delay the inevitable and "free" was eaten up by legal fees. 

When Mike finally was beaten, QSI came up with both the upgrade chips and a programmer that would make the upgrade a universal chip capable of having any sound file loaded into it or having the chip pre-programmed with specific sound files. Most problems with the Atlas and early BLI locos can be solved with the new chip. The original chips have a LOT of quirks and failures hanks to Mike. The chips are still available from different vendors and I think one is Litchfield Station. Tony's is another. They usually run about $30.00. Quite a bit cheaper than "removing the decoder" with another brand.

The suggestion on replacing the wheels is a possibility but after almost 25+ years of operation on both home and club layouts, I have not had to do that on any of mine yet. One small exception was the case where two of three consisted locos "de-consisted" at an op session and the operator only mentioned how bad they pulled his 15 car train. Naturally, they were both six-axle units and all wheels were flat-spotted after the one loco dragged them over the entire layout. Lesson learned, do NOT have consists coupled before the session.

Regards,
Nick Kulp


Jay Beckham
 

Nick

 

Were those OEM chips in the O Scale Atlas Gold series by any chance?

 

Jay

 

From: QSIndustries@groups.io <QSIndustries@groups.io> On Behalf Of Nick via groups.io
Sent: Saturday, September 12, 2020 6:05 PM
To: QSIndustries@groups.io
Subject: Re: [QSIndustries] 2010 Atlas Gold TrainMaster randomly stops and restarts

 

Hello "Loco loco" (Please sign your posts, it makes it a little more friendly)

The one small piece of information you have not provided is, have you upgraded the "crippled" chip in the decoder ? The original OEM chips in the first series of Atlas and BLI locos were victims of the Mike's Train House fake lawsuit where he falsely claimed to have invented features used in the decoders. He naturally lost the suit but only after forcing years of legal battles. During those years, QSI was forced to disable certain features of the decoders like BEMF, speed tables and more I can't remember. Since manufacturers had already started production on those locos, QSI placed "temporary" and removable chips in the decoders. The plan was to provide free upgrades to owners after the suit was settled. Unfortunately, Mike used years of legal tricks and false info to delay the inevitable and "free" was eaten up by legal fees. 

When Mike finally was beaten, QSI came up with both the upgrade chips and a programmer that would make the upgrade a universal chip capable of having any sound file loaded into it or having the chip pre-programmed with specific sound files. Most problems with the Atlas and early BLI locos can be solved with the new chip. The original chips have a LOT of quirks and failures hanks to Mike. The chips are still available from different vendors and I think one is Litchfield Station. Tony's is another. They usually run about $30.00. Quite a bit cheaper than "removing the decoder" with another brand.

The suggestion on replacing the wheels is a possibility but after almost 25+ years of operation on both home and club layouts, I have not had to do that on any of mine yet. One small exception was the case where two of three consisted locos "de-consisted" at an op session and the operator only mentioned how bad they pulled his 15 car train. Naturally, they were both six-axle units and all wheels were flat-spotted after the one loco dragged them over the entire layout. Lesson learned, do NOT have consists coupled before the session.

Regards,
Nick Kulp


kelly dorf
 

They looked OEM to me.  I am checking if we ever made an upgrade.

Kelly


PennsyNut <pennsynut@...>
 

Not a guarantee. But: After cleaning wheels and track, dab a bit of graphite every here and there and let cars spread it. Graphite helps electrical conductivity. Very little is needed. If you dab it and can see it, it's too much. This helped me on all locos. FYI My QSI is in the shop right now being checked. Seems to be a major problem, maybe cracked gear. But what I'm suggesting here had helped me tremendously with all locos.
Morgan F Bilbo, DCS50, SPROGIIv4, JMRI 4.20, Pennsy modeler 1952


Dave Horn
 

Thanks everyone for your suggestions.  Kelly is looking into the chip.  I look forward to hearing from him.  In the meantime here are some additional observations:

1) I ran the EL TM with a MRC 6 2.0 connected to the same loop that the NCE Power Cab had been connected.  It did not stall with the Tech 6.  My 2004 Lionel HO UP gas turbine (with 2005 updated chip) failed after 30 minutes of running with the PowerCab.  A BLI AC 6000 run with the PowerCab does not fail.  I'm starting to think it has something to do with the combination of a 2005 or 2010 QSI decoder (maybe others) and a NCE PowerCab (v1.65).  The unpredictability of it is frustrating, but the failure undeniable when it happens.  I wouldn’t care so much if it was just sounds cutting out, but the abrupt stop and start derails cars in tow, especially in curves. 
2) I've done the DCC all-CV's reset and the wand DC factory default reset several times. 
3) I ran it without the shell - nothing gets warm or hot.
4) I have not yet done the dark room spark observation, but it can fail 2, 6, 16, or up to 30 minutes once running.  I'm OK with running it that long to see if it fails, but haven't found the time yet to sit there in the dark staring at it until that happens.
5) There is only one engine on the track at a time.
6) I haven't yet tried the amp display on my PowerCab, but will.  Unfortunately the failure happens at any moment and very fast. I'd have to stare at the amp read-out with undivided attention and hope to see what value it registers the instant it happens without the means to trap or record it.  Like the spark test, not there yet.
7) No track wiring changes, all though I did inspect every connection since this problem occurred and even removed toggle switch so everything has a direct path from DCC system to tracks.  Still, the NCE PowerCab-QSI combination generates the problem.  
8) The PowerCab is at rest when the failure/stall happens.  I've moved, wiggled, and straightened the cable and jiggled both connectors while an engine is running.  I can't make a stall happen.  They do so under their own volition.
9) Not sure about spreading graphite on my rails. Might help with continuity, but no other engines stall or fail like my QSI decoder equipped engines do when run with my NCE PowerCab.
10) Replacing the wheels might help, but these have lots of silver plate and shine to them.  I have engines where that's worn off done to the parent metal (brass or copper?) and even they run OK although I'd rather they still have their NS.
11) CV11 to zero.  Can you provide more information on that?  Have you done so with a QSI decoder and it solved a problem?  What is its QSI indexed address and what is the factory default value?  Curious why you recommend this measure.  Kelly, are you familiar with it?  Thanks.   


Thanks again for you insights, questions, and suggestions.  Anything else please let me know.
 

Dave H.


Dave Horn
 

It is the original chip (2010 run of Atlas TrainMasters).  I provided Kelly a photo of the label on the chip.  He's having some laptop issues and will let me know more about that when those problems are resolved.

I set CV11 to 20 using Program On the Main CV11=20 on the PowerCab not with indexed binary values programming per the QSI 2006 Gas Turbine manual.  Does it work using PowerCab POM and the CV number instead?  The loco announced back to me “C V one one equals two zero” so I assumed it stuck.  Correct me if I'm wrong, thanks.  The engine ran several minutes and failed.

I set CV11 to zero.  The engine failed after a few minutes.

I set CV11 back to 1 (the factory default I believe), used PowerCab Option 1 to input the long address of #1860, and then Option 3 to Configure the loco with DC ops turned off and the long address activated.  The engine ran for 39 minutes without interruption and I ran out of time to run it any longer.  I hope to test it more tonight.  I don't count myself out of the woods on this yet, but I would like that to be the case.  If so, then I'll try this with my Lionel HO UP gas turbine and see if it stops it from abruptly stopping and starting when run with the NCE PowerCab.  Both locos as mentioned before work without interruption on my Tech 6 2.0 in DCC mode.   I'd love for this NCE ops issue to be resolved, but then also to know why the actions taken or changes made to achieve that were what was required.  Right now I'm pretty close to just trying everything suggested, but without a whole lot of understanding why the fix takes affect when (hopefully) it happens.

Thanks again for your help in the meantime.  Let me know if you have any further questions or suggestions or if I didn't answer any of your questions so far.  

I'll let you know what I hear from Kelly.

 


Dave Horn
 

The 2010 Atlas Gold EL TrainMaster cab #1860 and 2004 Lionel UP Gas Turbine cab #75 (with 2005 QSI updated chip installed) each finally ran for over an hour without interruption while pulling a 10 car train across speeds ranging from 1 to 126 speed steps.  With sounds and lights on, and some sounds activated occasionally.  I felt good about that.  A number of suggested measures had been tried up to that point without success.  

The last thing I did to both engines was the following:
1) CV11=1 using PowerCab, POM, option2 CV, not the indexed with binary values way in the QSI manual, but the decoder read it back to me
2) Long address set to cab # (7500 in the case of the gas turbine) using NCE PowerCab POM, option1 Address, long, enter
3) Address the engine by original (default address) 3 b/c while the PowerCab knows its new long address apparently the QSI decoder does not yet know to use it
4) POM, Option 3 Config, all defaults except turn DC off and activate the long address
From there -- no issues with either engine -- and they both failed previously more than once within 30 minutes or less, especially the TM, I just bought the GT.  

I can't say exactly why those actions resulted (so far) in the uninterrupted operation of both engines or whether it was any one step in particular or the combination of them all is that corrected the problem, but I was relieved and pleased to run each of them at least that long without issue.  They're both a lot of fun to operate -- sounds, lights, looks, smoothness, heft, etc.  I hope their good behavior continues.   

I have not heard back yet from Kelly, but look forward to in general on this issue as well as in terms of possible upgrades.  

A couple of additional search results I found helpful were:
https://groups.io/g/jmriusers/topic/sdn144ps_and_cv11/29708582?p= 
thomasmclae  02/11/19   #156830  
On the club layout, if a Lok does the start/stop dance, we do two things:
Turn off DC enabled option, and set CV11 = 0. [NOTE: I tried 0 and 20 and still had stop and starts, however, while still with address 3 and DC on]
For some reason, some decoders are fooled into thinking the DCC signal is DC. Maybe to the signal pattern, or the boost in power from the last upgrade. (EasyDCC system)
And the decoder refresh cycle is long enough that small values in CV11, like 1 or 2, timeout and the Lok stops. Then gets the DCC packet and starts, times out.....
CV11 is not a sound thing, although on some decoders the sound will go off if the Lok stops. 
As I typically look for CV11 on the special or decoder specific page.
THomas
DeSoto, TX

https://www.elmassian.com/index.php/dcc/specific-manufacturers/qsi-equipment/qsi-programming-tips  
Select off the list or scroll down to "First time set up of loco address" (for some reason I cannot cut 'n paste text off this page or I would have for you.)  There you'll see NCE PowerCab long address specific observations and instructions. 

Have fun running your trains -- uninterrupted.


Dave Hastings
 

Kudos. Great news, you did not give up on them. Glad to hear they both pull and you are having fun with them. Hope they work out for you in the future.

Dave Hastings


On Friday, September 18, 2020, 07:27:33 AM EDT, Dave Horn via groups.io <davidghorn@...> wrote:


The 2010 Atlas Gold EL TrainMaster cab #1860 and 2004 Lionel UP Gas Turbine cab #75 (with 2005 QSI updated chip installed) each finally ran for over an hour without interruption while pulling a 10 car train across speeds ranging from 1 to 126 speed steps.  With sounds and lights on, and some sounds activated occasionally.  I felt good about that.  A number of suggested measures had been tried up to that point without success.  

The last thing I did to both engines was the following:
1) CV11=1 using PowerCab, POM, option2 CV, not the indexed with binary values way in the QSI manual, but the decoder read it back to me
2) Long address set to cab # (7500 in the case of the gas turbine) using NCE PowerCab POM, option1 Address, long, enter
3) Address the engine by original (default address) 3 b/c while the PowerCab knows its new long address apparently the QSI decoder does not yet know to use it
4) POM, Option 3 Config, all defaults except turn DC off and activate the long address
From there -- no issues with either engine -- and they both failed previously more than once within 30 minutes or less, especially the TM, I just bought the GT.  

I can't say exactly why those actions resulted (so far) in the uninterrupted operation of both engines or whether it was any one step in particular or the combination of them all is that corrected the problem, but I was relieved and pleased to run each of them at least that long without issue.  They're both a lot of fun to operate -- sounds, lights, looks, smoothness, heft, etc.  I hope their good behavior continues.   

I have not heard back yet from Kelly, but look forward to in general on this issue as well as in terms of possible upgrades.  

A couple of additional search results I found helpful were:
https://groups.io/g/jmriusers/topic/sdn144ps_and_cv11/29708582?p= 
thomasmclae  02/11/19   #156830  
On the club layout, if a Lok does the start/stop dance, we do two things:
Turn off DC enabled option, and set CV11 = 0. [NOTE: I tried 0 and 20 and still had stop and starts, however, while still with address 3 and DC on]
For some reason, some decoders are fooled into thinking the DCC signal is DC. Maybe to the signal pattern, or the boost in power from the last upgrade. (EasyDCC system)
And the decoder refresh cycle is long enough that small values in CV11, like 1 or 2, timeout and the Lok stops. Then gets the DCC packet and starts, times out.....
CV11 is not a sound thing, although on some decoders the sound will go off if the Lok stops. 
As I typically look for CV11 on the special or decoder specific page.
THomas
DeSoto, TX

https://www.elmassian.com/index.php/dcc/specific-manufacturers/qsi-equipment/qsi-programming-tips  
Select off the list or scroll down to "First time set up of loco address" (for some reason I cannot cut 'n paste text off this page or I would have for you.)  There you'll see NCE PowerCab long address specific observations and instructions. 

Have fun running your trains -- uninterrupted.