Weaver Geep, DCC running problem - now sorted.

[F-UnitMad]

I've posted this Topic in the 'DCC Help' Section, but thought I'd highlight it here as it's an American loco so might get more relevant comments. I'll post the link to the other thread at the end as it has photos added.

 

This is a real mystery, this one....

I have a Weaver O Scale Geep, 2-rail, Pittman motor, shaft drive via delrin chain on a tower to rear axle, then shaft drive to all axles.

A few years ago I fitted a 'boosted' Soundtraxx Tsunami 3amp dcc decoder to it, and all was well and good until last year, when the loco started behaving strangely - when the speed was set, at anything above speed step 1, the loco would get to that speed, and utterly ignore any further commands, like slowing down, stopping, etc. It just kept going regardless.

I thought maybe the decoder was at fault - I've found Tsunami's to be fussy things, liable to stutter at random, on clean track that other decoder-equipped locos go through no problem. 

So I've replaced the decoder with an ESU Loksound L, again 3amp rated.... and the same 'runaway' problem is occuring. While "in between" decoders I wired it for DC and tried it on analogue - no problems at all. I've checked the wiring and pickups for any sign of breaks or shorts  - all looks ok, and it doesn't trip the system anyway, which is NCE Powercab with NCE 5 amp  booster. Cleaned the wheels, including the rear faces that the pickups contact.

Then I've tried it with a similar motor (Pittman). Same runaway issue, so it isn't a motor fault, BUT - here's what's really mystifying - with the drive shaft disconnected, i.e. no load on the motor, it behaves itself ok. Starts, stops, reverses, all according to throttle settings.

Re-connect the drive shaft, and it misbehaves again. The fact it's done it with two different decoders, that were both new, and have even had CV8 reset by me just in case, leads me to say it's not a faulty decoder either, but clearly something is happening when the motor is under load that is scrambling the decoder signals. The drive train does not bind or catch, there's some slack in it, and with everything connected up it all turns easy enough by hand.

I'm utterly at my wit's end with this issue. Anyone come across anything like it before??? Or any ideas what might be happening??

 

Edit:- Link to DCC section post - 

 

Edited July 19, 2020 by F-UnitMad

[jasond]

Sorry Jordan, I promise I will may be a DCC user soon, but I'm ready to rise to the challenge ... so I asked Google and one of his links was a name I recognised:
https://sites.google.com/site/markgurries/home/dcc-problems/decoder-problems/decoder-loss-of-control

 

It may or may not have a useful clue ..  shut up now Jason.  Solve it and become our resident expert ... please!

Jason

[trevorsmith3489]

If the loco works OK and you have tried 2 different decoders -

 

have you considered your NCE Power cab could be the culprit? Perhaps a reset might be tried.

could the loco address be part of a consist?

Have you tried removing the booster from the system, and just running two wires from the NCEPowercab to a short length of track to eliminate a track fault/booster fault?

Does the loco work OK upside down on a cradle, ie eliminating any load

How does the loco perform if you use address 3?

 

Trevor

[long island jack]

May be when the drive shafts are fitted'it lets you start, but when the decoder senses the load, it's may be drawing more amps,than decoder likes,which shuts the decoder down, and it remembers the last setting,and keeps on going,just a thought!

 

 i use to have similar problem with cheap chips in HO, try disabling dc in cv29

[doctor quinn]

One of our members had a similar problem with a pair of Heljan Class 20s, they would start under control but then not response to commands. He fixed the problem by setting them to DCC only.

[GP9u]

Jordan,

 

I think Jason and jack are on the right track.  (I'm a poet but I didn't know it)  

 

I have a few questions for you:  What's your track voltage set at?  And does the Pittman have a "1" or a "2" as the third digit in the model number on the label such as 8514 or 8524?  Also is the Loksound "L" a LS5 or a V4 or a Select? And what gauge wire is your track bus and feeders? Lastly, I assume you don't have any cracked sprockets in the drive, i.e. it's smooth running with no jerks or jams?

 

If the track Voltage drops below about 14V under max load at furthest distance from the booster I've found you can have loss of control issues with NCE and Loksound (not sure about Tsunami).  The older Pittman XX1X motors have poor magnets and can draw a lot of current as they get older, spiking up to 8A under load. The XX2X models have better magnets and are a lot more efficient. 

 

Also you need to make sure you have DC conversion and Railcom off in the decoder as DC on will allow runaways if the DCC signal stops but power is still present, and Railcom can interfere with some DCC systems.

 

I hope there's something here that helps!

 

FWIW I sold off all my Pittman motors and replaced with Maxon coreless. Way more efficient and I get to use HO-sized  decoders leaving room for much larger speakers.         '

 

Pete

   

[F-UnitMad]

Thanks everyone.

Trying to cover most points raised:-

No it's not the Powercab. Everything else works fine.

This loco has never been part of a consist, and a CV8 reset has not solved the issue.

I always set CV29 with DC mode off, so it's not that.

It's a Loksound L Select, but I was getting the same issue with a Soundtraxx Tsunami ( a "boosted" 3amp one from Protocraft in the USA).

 Both motors I've tried are Pittman 8514, built in 1990 & 1995 respectively!!

I have no way of measuring track voltage, and I don't know if the output is adjustable with NCE, I think not.

The motor works fine & behaves properly with the drive shaft disconnected. With the drive connected, it plays up as described in the OP.

 

From replies here and on Western Thunder it's looking like the drive line is placing excessive load on the motor, affecting the decoder.  Pete (GP9u)'s post is of particular interest. As well as a serious dismantling & lubrication of the drive line, I'll have a look for Maxon motors. Any links or pointers to a direct replacement for the Pittman 8514?

Thanks chaps!!

[Regularity]

Canon FN30 motors are often mentioned on MRH: similar size, but only a quarter of the power draw.

Rod Miller has them Listed for $40 each plus shipping. You can probably find nearly new versions on eBay. (E.g. a box of 14 for about £160. Get together with some friends!)

 

Don’t know of a UK supplier, but you know what to look for!

Edited May 15, 2020 by Regularity
Typo

[GP9u]

Hi Jordan,

 

I've been using Maxon RE25 118751 as a direct replacement for the Pittman 85xx series. The torque is comparable for about 80% reduction in current draw.  I got mine surplus from Micro-Locomotion but no longer an option as the proprietor Eldon Shirey sadly passed a couple of years back. These motors are available online as surplus. Sometimes they have been relabelled if they were originally sold with a gearhead or other adapter.  See pics.

 

I just bash a short length of 4mm OD brass tube onto each end of the shaft to fit the ID of the P&D/Weaver couplings.

 

See also the MP15 AC chassis "test bed" pic with the Maxon RE25 in place of the original Pittman. The overall length is about the same, OD a bit smaller. This allows room for the HO decoder on top, leaving room in the shell for a Tang Band 1931s speaker.     

 

These motors are smooth and quiet with some tuning of the Loksound motor control CVs.  I use HO-sized LS decoders with them and have no issues at all in the last 5 years.

 

Pete

 

 

 

[APOLLO]

Hi - you mention drive it could be 

2 hours ago, F-UnitMad said:

 

The motor works fine & behaves properly with the drive shaft disconnected. With the drive connected, it plays up as described in the OP.

 

From replies here and on Western Thunder it's looking like the drive line is placing excessive load on the motor, affecting the decoder.  Pete (GP9u)'s post is of particular interest. As well as a serious dismantling & lubrication of the drive line, I'll have a look for Maxon motors. Any links or pointers to a direct replacement for the Pittman 8514?

Thanks chaps!!

 

I don't use DCC but have had problems with one or two of my Weaver O locos, narrowed it down to a split nylon gear on the tower drive. I got replacements from P&D model shop in the USA and this cured the problem (sticky hesitant drive)

 

https://www.pdhobbyshop.com/Store/index.php?cPath=23_26_37_104

 

More info here

 

https://ogrforum.ogaugerr.com/topic/weaver-rs3-question

 

https://forum.aorailroad.com/t/p-d-weaver-gear-towers/108/21

 

It seems the original nylon gears were made from a type of nylon that was prone to splitting. If you inspect the gear with a magnifying glass you can see the crack (if there is one) - even a small crack will ensure poor performance.

 

Brit15

 

 

 

[F-UnitMad]

A bit of info on power consumption.... I did some tests with the Geep on the 'basic' NCE Powercab setup, so I could read the Amps.

Just sitting with sound on, .11 amps.

Low speed steps, varying up to about .45 amps.

At speed step 15, getting close to 1 amp.

 

At this point the loco was behaving itself and responding to the throttle properly. So I went for it and tried step 28 - the amps went over 1.3, and control was lost - no throttle resonse, no sounds like the horn, and it's a good job I have a roundy-roundy!!

I'm well aware that this was exceeding the amps the basic Powercab puts out anyway, but does seem to confirm that the Pittman motor is too amp-hungry under load, & needs replacing.

[long island jack]

Maybe that Tsunami isn't that fussy after all!!! lol

 

Looks like a new motor will be on it's way!

[F-UnitMad]

My experience of Tsunamis hasn't been great in either HO or O, to be honest. They've been 'fussy' in whichever locos I've fitted them in, seeming to require ultra-clean wheels and track, and even then sometimes cutting out or stuttering for no apparent reason.

My SW1200RS has one, though, and at least this test has shown this 'boosted' 3amp Tsunami isn't faulty as I thought it was, so might well get fitted to my SD40, (same Turbo 645 engine) which currently just has an NCE 408 non-sound decoder in it. Every cloud, etc....

[GP9u]

Jordan, a couple of observations if I may:

 

Is that Tsunami a v1 with the 3.5A add-on module?  If the decoder can handle 3.5A but the DCC booster can't deliver that, maybe the booster's Voltage drops off as it tries to deliver more current.  The booster might not shut down like it would for a short because it's seeing a gradual rise in current draw, not a spike (boiling frog syndrome!). 

 

That drop off could make the DCC signal degrade to the point where the decoder can't read it any more so it doesn't know what to do, even if the DC running option is turned off.

 

So, two things I would do next: 1) purchase a RRAmpmeter and 2) add keep-alives to all your DCC decoders to cure the "fussiness".

 

The Amp meter is very useful in O scale where we have to gamble with motors of unknown performance and potentially large current draw a lot more than in HO or N. It can also measure DCC track voltage quite accurately which is very helpful in testing.

https://tonystrains.com/news/rrampmeter-by-dcc-specialties/

 

Over time I've found that adding keep-alives is only way I can enable multi-hour ops sessions with no LED flickering or sound dropouts ever. This goes hand-in-hand with feeders to all track sections and DCC-friendly turnouts, no exceptions!  Some good info here:

https://sites.google.com/site/markgurries/home/decoders/keep-alive-compatibility

 

Connection info for various decoder types here:

http://www.members.optusnet.com.au/mainnorth/alive.htm

 

Tedious work but a one-time fix and well worth the effort imho.       

 

Pete  

[F-UnitMad]

4 hours ago, GP9u said:

2) add keep-alives to all your DCC decoders to cure the "fussiness".

I've only really experienced 'fussiness' (I can't think of any better description!!) with Tsunamis. Loksound, NCE, Lenz, TCS, QSI have all performed well - still need clean track & wheels of course, but not prone to stuttering at random like a Tsunami, and the Tsunami in this Geep did have a KeepAlive fitted - a big TCS one, but it didn't overcome the problem completely.

I've no idea of the booster with it; Protocraft in the USA do it as a conversion of a standard 1 Amp decoder, with extra gubbins & a heat sink. I wired & fitted it as per their instructions.

The layout is all wired to the bus, hand-built switches with Tam Valley Frog Juicers, so all 'good practice' been followed. In general everything (except this loco now!) performs well, and DCC gives an operational flexibility that would be almost impossible to do with analogue, or at the very least would involve some very complex wiring & cab control.

The next step for this Geep is definitely swapping out the Pittman motor for something else, funds notwithstanding, and we'll see what happens with the Loksound then.

Edited May 18, 2020 by F-UnitMad

[GP9u]

I had one of those Protocraft upgraded Tsunami 1s and it worked great in an Atlas China drive GP35 for several years.  I did add a TCS KA to it also and no issues on my old Digitrax system with 8A boosters (before I started using coreless motors)!

 

But that Pittman can draw a lot more current than an Atlas dual-motor drive for sure.  I just noticed a thread on the NCE-DCC group that might shed some light:

https://groups.io/g/NCE-DCC/topic/trip_current_for_a_powercab/74262809?p=,,,50,0,0,0::recentpostdate%2Fsticky,,,50,2,0,74262809 

 

As you say, you have everything else sorted, so maybe try a Maxon RE25 118751 or similar next. Btw I was pleasantly surprised that used Pittman 8514 motors are still in demand and I was able to sell mine for about half the cost of a surplus Maxon. 

 

Pete

 

[F-UnitMad]

On 15/05/2020 at 18:29, Regularity said:

Canon FN30 motors are often mentioned on MRH: similar size, but only a quarter of the power draw.

Rod Miller has them Listed for $40 each plus shipping. You can probably find nearly new versions on eBay. (E.g. a box of 14 for about £160. Get together with some friends!)

 

Don’t know of a UK supplier, but you know what to look for!

A bit of an update on this Thread. 

Even 'used' Maxon motors seemed to be rather over my budget, so I found a US Ebay Seller with new/unused Canon FN30 motors for sale. He had a box of 14 available (Edit - must be the one you mentioned, Simon!!) but also just 4 in a box as a seperate listing.

Long story short is that including everything (shipping etc) they've basically cost me £20 each; enough for my two Weavers, my Red Caboose/Atlas GP9, and 1 spare. They arrived today.

Shorter than the Pittman, but the flywheel & coupling have fitted just right on the shaft.

Wired it up, & off we go. NCE Powercab reports .11 amps at rest, and no more than .65 or so amps at Step 28, and there was no unresponsive 'runaway' behaviour; normal response to the throttle from all speed steps up & down.

 

Fortunately despite the shorter motor length it still sits on the 4 molded spigots on the chassis for the Pittman motor, and the connecting shaft from motor to tower is extendable which has come right in handy!! I just need to fit a cable tie to hold the motor in place, tidy up the wiring, and refit the body - then test again just to be sure!! But it looks like this was the cure to the problem - the Pittman motor being just too Amp-hungry.

Plus I realised my NCE Booster is the SB3a, not SB5a, so only 3 amps, not 5!! D'oh!!

I don't think it's made any more.

Top speed isn't very fast, but that's also because another modification I made was to replace the 12-tooth cog on the bottom of the gear tower with an 18-tooth one. Boy this loco can crawl, now!! Next step will be to work out if I can adjust the speed steps that the throttle sound notches up at, as now it seems like it's in Notch 5 or so and still crawling!! There is a manual notching option but as usual the Loksound Manual is very unclear on how it is actually activated.

Anyway happy bunny and thanks to Simon especially for the heads-up on Canon motors!!

Edited June 2, 2020 by F-UnitMad

[Regularity]

Glad to have helped, Jordan.

My only worry is that with the extra motor, you may buy another set of trucks etc and build a CF7 around it...

(They are growing on me in a bauhaus sort of way. Comparing them to some of the latest imports to this country helps!)

[F-UnitMad]

8 hours ago, Regularity said:

 

My only worry is that with the extra motor, you may buy another set of trucks etc and build a CF7 around it...

 

[DanielB]

Thanks for the update on this Jordan. I've recently fitted a loksound V4 in my Weaver RS3 and it's not playing nice AT ALL. I'm beginning to wonder if it's the same root cause as your issue here.

[F-UnitMad]

9 hours ago, DanielB said:

Thanks for the update on this Jordan. I've recently fitted a loksound V4 in my Weaver RS3 and it's not playing nice AT ALL. I'm beginning to wonder if it's the same root cause as your issue here.

I'd say now after this, that it's highly likely!! Do you use the NCE Powercab, or any other way of finding out what Amps it's drawing? The thing was, even when the Pittman motor wasn't drawing big amps, it was clearly affecting the decoder, as I said at the start of this thread, anything over speed step 1 rendered the loco uncontrolable.

[DanielB]

I do use a powercab yes, though other than programming CV's, I'm rubbish with using it beyond driving locos

 

Oddly, my Atlas GP35 works absolutely fine with the same decoder type. I'm assuming Atlas locos use much better motors?

Edited June 10, 2020 by DanielB

[F-UnitMad]

To get the Powercab to show amps, do this:-

a) In the lower left corner of the Power Cab keypad press the Program / Escape  (PROG/ESC) key 6 times.  You are looking for "SET CAB PARAMS"  

b) Press "ENTER" key.

c) Display will now say: "SHOW TRK CURRENT 1=Y". 

d) Press "1" key.

e) Press PROG/ESC key 2 times to get back to the main screen.

 

The clock display will be temporarily replaced by the current meter display.

It won't do this if you use a booster, it needs to be using the 'basic' setup, with the flat 6-wire cable. 

 

The Atlas motors draw less amps than the Pittman in the Weaver, and if you wired them in series they'll draw even less. My GP35 uses a TCS Wowsound decoder, the 'HO' one, i.e. about 1 amp. 

[F-UnitMad]

As a late post-script to this thread, today I fitted the original Protocraft "Boosted" Tsunami chip to my Atlas SD40, which of course shares the same prime mover as the GP40. The chip has performed just fine with the Atlas motors wired in series, further confirming that the 'runaway' problem that developed with this chip in the GP40 was down to the power draw of the Pittman motor, not the decoder.

 

[jasond]

Have you had the time to fit one of your Canon motors yet?