Skip to main content

Max, after another sleepless night worrying about your predicament     I am feeling like the sounds in your video mean DCC "track power" or some incorrectly processed power is going to the motor. The frequency of the buzzing and the way it steps down in line with throttle step increases is exactly the same as my Pittman locos sound running on DCC with Digitrax "zero-stretching fake DC" before decoder install. Bear in mind also that the 20kHz and 40kHz DC motor drive options are out of our hearing range (your dog might hear the 20kHz) so this is definitely not that type of buzz that we're hearing.   Couple of things maybe worth trying:

  1. I think it's key that the Loksound guys hear your buzzing. It's nothing like I was expecting to hear from variable PWM during back EMF optimization. I expected to hear a lower frequency wum-wum-wum kind of pulsing buzz that would slowly and smoothly speed up and fade out as the loco gathered speed. If they are thinking that too, then when they hear your actual buzz they might realize it's a different problem altogether (pinch of salt with this as I may be way off base here)
  2. I recall you said the decoder runs the motor fine on the tester? So... one thing you could to try is take track power from the track to the tester, through the decoder mounted in the tester then to the motor in the loco. If there's a wiring issue that's bypassing the DCC motor control somehow this would show it. It would show if the ESU "decoder cradle" is somehow at fault if your using a different one in the tester than the loco. 
  3. Depending in outcome of 2, if the decoder works properly in the tester then you likely have a wiring issue in the loco. If not, then try bypassing the decoder setup completely with some test leads direct from track to motor. However, this assumes your DCC has a "run DC locos" feature and it's enabled. if you hear the same stepped buzzing then maybe there's a decoder setting or fault bypassing the full DCC motor control system.

Sorry this is clutching at straws and I really hope the Loksound support guys can help.

Agree with Andre re the truck issue. Normally it's an insulated wheel not pressed dead square onto the axle. But if not, then it can also be caused by the drive tower tilting back and forth under "torque reaction" where the UJ yokes aren't 100% square on their shafts (it's hard to get them dead square as their bosses are so short).  The truck equalization should be the final cure for this symptom but in practice the side frame snap-in pins are such a tight fit that these trucks rarely equalize freely enough. Hence the need to keep the drive shaft slider freely movable to avoid wheel lift.Yet another part that needs blueprinting and ideally some lube (although I'm not 100% sure what to use).

As I may have mentioned, apart from all this, it's a great drive system...     

Keep at it!

Pete 

Hi Guys

If you look carefully (as I finally did), in the video you will see that every axle is out of alignment.  On the track, the mass of the loco tends to hide the rocking.  It's the moaning noise that is the major problem.

The motors do not make the noise until they have the load of the drive train on them.  As I said, even the Atlas makes the noise - but it's so soft that it's easily masked by the engine sounds from the speakers.

I'm beginning to think that we are pushing the boundaries by trying to drive these big DC motors with a decoder.  My strong impression is that Pittman motors are at the bottom end of the quality scale with regard to their design.

I have to say that even in HO, there is really only one brand to choose if you want good performance - Atlas.  Remember that I'm talking computer control here.  On my HO layout, I have Rivarossi, Proto 2000 and Atlas locos.  Only the Atlas will run without cogging.

I could start chasing a motor which will have the specs to handle DCC (pulse modulated supply), but frankly I think it's a waste of time.  The loco runs fine for general modelling with the BEMF turned off.  I'm sure that I can find a home for it with someone who isn't running computer control.

Thanks to everyone for their input.

I am not an expert but I use a PWM controller that caused my motor hum and heat up.

the motor was an old pittman without BB. I blame at the first until I troubleshooting  step by step founding the controller was not good.

i tested with pure dc and the motor wasn't hum and  heat.

Andre.

 

Hi Andre

I think you're right.  PWM is the way decoders drive the motors, and the Pittmans just don't like it.  The Atlas don't like much, either, but it's bearable.

What is interesting is that I have an Aristo G scale RS-3 which I drive with a Loksound XL without any of these problems, so it's not motor size - more motor quality.

When I started into O, my friends told me that O scale is like the Wild West of modelling.  It's good being back on the frontier. 

Cheers

Max, I gotta say it seems a bit out there that Loksound is using a variable PWM and Back EMF drive that is so different from Tsunami and TCS-WOW that it can't drive these motors quietly like the others do. I am still suspicious of the type of buzz we're hearing on your GP38. I have never heard any buzz like that on any Atlas, MTH, Yoder or Pittman motors except without a decoder on DCC. 

Looks like no Loksound for me, except on my Yoder 44 tonner where it's fine. 

Sorry we weren't able to fix it Max.      

Interesting Max. I realized that the only combo I have yet to try is Pittman with Loksound. I got to thinking that maybe the Euro DCC makers (perhaps Lenz, ESU, Zimo) do use a different motor drive approach. There are a lot of modern high-quality motors including coreless such as Faulhaber, Maxon etc. in the hobby over there These motors use a lot less current and have much lower rotating mass and presumably inertia. So maybe it's not such a stretch to think a drive optimized for them might not be a great fit with old school high-amp motors like Pittman. Even a more modern Pittman with ball bearings and Alnico magnets would present quite a different load I would think.

I also Googled up some "Loksound buzz" videos and I did find an HO one with a buzz that sounded like what you have. Anyhoo, that's more than enough fretting and BSing from me, let's wait to see what the actual experts say!     

Pete

Last edited by Pete M

Both Dave and Mark have responded.  The decoders have adaptive regulation frequency, which is the number of times/second the BEMF is sampled.

Unfortunately it's not possible to turn it down enough to cancel out the poor build of the motors.

The noise is directly proportional to the tolerances in the motors.  Clearly Atlas motors have a closer tolerance than Pittman, so the buzz is very soft compared to the Pittman.  It's still there, but the decoder sounds drown it out.

I think that I've come to the end of the line with this.  I don't propose to waste any more time or money on this loco.  I'll switch off the BEMF and then massage the performance CV's to get the nicest effect I can manage.

Once I get the layout built and the computer system installed, I'll find out if it can cope with being driven hard by the software, or if it begins to skip over commands when the whips are cracking.

Cheers

Understood Max. I have reached the conclusion that trying to make modern decoders control old technology motors is barking up the wrong tree. I have been looking at replacement motors, especially coreless, such as those now being used by Jay Criswell and others.

Back in the mid '90s before I left the UK for Canada, coreless motors were common in 7mm British steam models. And way more common among European modellers. Finding motors with metric dimensions that are close enough to retro fit into a Pittman/Weaver/P&D drive line seems to be the one of the few issues. Torque, power consumption, control resolution and mechanical quality is already where we need it.     

Jay offers a range of complete driveline replacements with a neat belt drive system and multiple motor and gearing choices. Sadly, for now I don't have funds to have him do those complete upgrades. My hope is to find an "almost drop-in" motor to replace the Pittman.

Hopefully you can get decent running characteristics from your GP38 without the back EMF. The NCE D408SR in my P&D F7 doesn't have back EMF and seems to run pretty well after some basic CV tweaking. My 1/8" top sprockets just arrived for my tower kits, so I will try to get that done and post the results.

Thanks again for sharing all your work. I have learned a lot!

Pete 

 

I have five Weaver locomotives from different era. I even have an operable center drive. IMHO that was the best and simplest. Most people thought it was too noisy so Weaver went through several mods to resolve the noise issue. Frankly, the end tower version was somewhat quieter but had its own problems. The chains tend to stretch in time and sprockets need to be removed. Weaver's biggest failure was in not installing an idear gear on the tower so the chain could be tightened SLIGHTLY as it stretched. If you tighten the chain too much and fail to lube the shaft, the hole through the tower will heat up and eventually seize the gear shaft. You then have to use a #7 or #8 drill bit to ream the hole in the upper tower. In all honesty, before Weaver went to the China drives, the old chain drives were a pain in the a__! Thank God P & D is around to supply parts for older Weaver locomotives.

Last edited by Tom Platten

Interesting, Tom.

From my observation, the P&D replacement parts are all Weaver made.  The axles are all bent (like the OE ones), and the towers need Kelly's replacement adjustable towers (and larger bottom sprockets), to make them half decent.

And don't get me started on the wretched Pittman motors!

Let's face it, Weavers are entry level locos.  They lack important details, like cut levers, MU hoses, windshield wipers, et al.  So their drives are unsuprisingly, very basic.

I need to keep in mind, that O scale is still in the dark ages where everything is predicated on DC operation.  Even the beautifully detailed Atlas has to have the motors rigged in series to get any sort of acceptable perfomance on DCC.

I have to say that I am enjoying my transition to O scale.  It's like the early days when HO went DCC.  Unfortunately O doesn't have the economies of scale of HO, so we have to expect it to be like the Wild West - for the time being, at least. 

Cheers

MaxSouthOz posted:

Interesting, Tom.

From my observation, the P&D replacement parts are all Weaver made.  The axles are all bent (like the OE ones), and the towers need Kelly's replacement adjustable towers (and larger bottom sprockets), to make them half decent.

And don't get me started on the wretched Pittman motors!

Let's face it, Weavers are entry level locos.  They lack important details, like cut levers, MU hoses, windshield wipers, et al.  So their drives are unsuprisingly, very basic.

Cheers

I guess I got lucky, then. My recent purchase of this same model runs very well. It has the same Pittman as is in my Red Caboose GP9, and shows a similar performance. Smooth and quiet. 

These Weaver diesels remind me of the Athearn "blue box" engines of the past in HO. Solid, good running (usually!), and a good foundation on which to detail as much (or little) as one might like. I personally prefer this "platform" over a more expensive offering that features a lot of bells and whistles I may not need or want. This way, I can enjoy doing the work myself!  

 

Mark in Oregon

Ah!  That's it.  Mine ran fine on DC as well, Mark.

As soon as we went to DCC, it ran less well. 

We're still working on it, but apparently the Pittmans are at the bottom of the specs table and could be the root of the problem.

I'm going to runs some more tests, but it's Christmas Eve here and the extra test decoder hasn't arrived from the US; so it looks like it will be some more days before I can play with it again.

I'll just have to content myself with the egg nog. 

Hi... your locos look excellent. Those of us who were long-time HO-scalers before we switched to (or back to) O scale have a somewhat different approach to modeling in O scale. Not better or worse, just different. But hang in... after awhile you’ll get the bugs worked out and then enjoy the heft and sounds of O scale on the rails. I have several Pitman-driven Diesels... the motors per se never have given me any problems. Often it’s the drives or electronics which need help. In my HO days, all of my Diesels were powered with Pitman DC motors from open-frame permag DC60 to DC90s, and all were hooked up with the old Hobbytown Centri-clutch drives with multi-flywheels. Real clutch-drives were fantastic, could crawl like an ant, and at high speed would coast for several feet if the power was shut down. A real kick. I have never found anything that good in O scale. Probably because of the much heavier weights involved... plus electronics and simple flywheels took over for the coasting effect. I just posted a video on this forum showing my on-board battery power for O scale locos. I think you will find it in the recent postings column.

 

Best wishes, and Merry Christmas!

~Andy R!

Last edited by Ironbound

Thanks, Andy.

I've lived (just), through the transition from DC to DCC in HO.  It's all very familiar.

Just to beat myself up further, I run my layouts by computer control.  I'm not sure how I would do it using RC. 

It's a very enticing concept, though.  Give me time.

One thing about banging one's head on a wall.  It feels great every time I stop for a break. 

Merry Christmas. 

I finally got my first Finescale 360 tower into the P&D F Unit. This included dropping the top sprocket from 10T to 8T. It has the brass trucks, the full "Bob Sobol inspired" upgrade package as well with 20T lower sprocket. I was able to get the top shaft almost up to the motor drive shaft C/L by adding 2 chain links. The chain tension is very easy to adjust with the cap screw and sliding upper tower.Way better than before and at least it's parallel to the lower shaft!

Also the UJ pivot can easily be aligned directly above the truck pivot, and the top and bottom sprockets can be adjusted to perfect alignment. I used one P&D brass washer to align the top sprocket. 

The decoder is a D408SR which doesn't have back EMF. I would say the noise is down by around 75% at speed, and the starting off and slow running is 50% better.

 

Link to after video: 

 https://youtu.be/dCtO6spO-rs  

Max, I can't hear any motor hum with this setup, so hopefully your Loksound with back EMF turned off will deliver at least as good results, bearing in mind the D408SR is a "previous generation" decoder. I am now looking at coreless motors e.g. Faulhaber, Maxon, Portescap. I used Portescap in British 7mm modelling back in the early 90s before DCC. But the early Digitrax decoders were too low a drive frequency for coreless so I used the provided Pittmans up to now.

There's a lot of good info out there so I'll let you know what I end up with.    

Learning all the time!  

 

Last edited by Pete M

Thanks, Pete.

I'm still hammering away at ESU.  They released the L series decoder specifically for O scale and it's come up short.  They close ranks when there is a design problem and attack anyone who dares to question them.

Their HO decoders are tops - as are the XL for G scale; they've just plain got the L series wrong.

If you find a good quality can motor with the right size shaft, please let me know and  I'll have a crack at it.  I still think my GP38-2 can be made into a good loco.  I'm with Mark on this.  The basic model lends itself to super detailing.

BTW, I can't see your video.

Cheers

I agree the GP38 can be a good loco. I did one in CP with grab irons, winterization hatch etc. and it looks decent. It's not 100% accurate but plenty good for me. If only I could make it run smoothly...  

For some reason I can't get the video to embed after the forum upgrade, but I put the link above. Hope it works for you now!

I am talking to "the Motorman" via ebay who seems to have lot of options for Faulhaber motors that many others have already used to replace the Pittmans so I'll keep you posted.

It's a shame about Loksound's attitude. Matt Herman gets up here quite often to visit Rapido and George's Trains. He seems like a great guy so maybe I can engage him next time. 

 

    

Yes.  Matt is a good bloke.  In understand that they are protective, but some of us older blokes have been with ESU since before Matt, so we've seen ESU for a long while.  It might be a German thing.  It will pass, as always.

I got the link to open.  The boss on the flywheel looks a bit eccentric, but it's all running nicely. 

 

Oh well, like you say, it'll pass. By the way, I have been highly impressed with the HO TCSWOW Diesel decoder in my Atlas O SW9 and they have an "O Scale" offering now, with built in keep alive, that I want to try. But they don't have the right sound set for the GP38-2 yet.  Can't win 'em all I guess...

Yeah, that UJ boss inside the flywheel is a pain. It uses a single grub screw to tighten it as it doesn't push onto the motor shaft quite enough. So when I tighten the grub screw it pushes slightly off centre. I spent ages trying to hold it off-centre the other way, so that it would be dead-centre when tight, but this is a close as I got.

Model Railroading is still considered fun in some quarters, right?   

 

LOL Max, now I get it! 

OK... Game on:

Next in play are 3 x Red Caboose GP9s to get the full Finescale 360 chassis and tower treatment plus Bob Sobol-style pickups. One, a high-hood has the same D408SR/DSX setup so I'll go with that first. The other 2 are CP chop-nose conversions on P&D trucks. I'll get them running then decide about decoders once I figure out the coreless motor thing.

At least I feel like I know what I'm getting into this time around and I have all the drive parts in hand. I think the Finescale 360 tower kit plus the right coreless motor and decoder with keep-alive will solve this drive for modern DCC/sound.

Silver lining... 

 

The Red Caboose GP9 has excellent detailing. The original kit was a GP9 high-hood out of the box. The ones I have are from the kits which have amazing but very fragile details. Grab irons, lift eyes hoses etc. are all plastic so they easily get knocked off. But everything you need to swap out for metal wire and cast parts are readily available from P&D, Des Plaines etc., all the usual suspects!

The original kit used the old Atlas F Unit double-ended drive with a twin-flywheel motor (you guessed it - a Pittman)! Very nice slow running on DC light engine. But those old Atlas "top-pivot" trucks were not good for keeping wheels on the track under load. So I eventually figured out that the P&D upgrade kit was the way to go. Then I found the Des Plaines brass chassis plate, and now the Finescale 360 version. The Des Plaines brass one is hard to get now. It had the advantage of weight over Kelly's kit. But I like Kelly's better as it comes with chassis rails machined in and milled pathways for ditch light wires that I need. Plus he supplies weights so it's all good.

That said, the way I understand it is that Atlas bought the tooling for the excellent body shell and maybe some of the frame too. Then they tweaked it to work with the dual motor China drive. I think they modded it to make a GP7 also (anyone please correct me). I have never seen one so I may be way off base here. Not sure if they had to compromise handrails or much of the body to work in 3-rail and the dual motors.

So if you find an original kit, I would recommend the P&D brass trucks, Finescale 360 chassis, Weaver/P&D drive with Kelly's tower upgrade and coreless motor TBD as soon as we figure out the best one. I think P&D or Des Plaines bought up all the remaining body kits which were left over without the driveline. So if you can find one and add the above items you'd have a "state of the art" Red Caboose GP9. 2 of mine were chop-nosed back in the UK 20 years ago by a mate with way more styrene-wrangling skills than I. Soon after that, (of course!) Des Plaines made a window surround for the chop-nose version which afaik they still offer.       

Hey, it's a sport, right...     

     

Interesting that the Red Caboose Geep 9 has come up. I spent a good part of yesterday working on mine.

I got it off eBay a few years back; AT&SF freight colors. It came assembled, but the original builder was less than awesome as far as glueing and "color matching": blue hand grabs on the yellow ends, yellow lift rings (most were broken off) on the blue roof... you get the idea. Anyway, I ordered some brass detail parts from P&D, removed and replaced the "offending" pieces, and it now looks so much better! I am having a hard time matching the Santa Fe blue: any suggestions?

It has always been a great running engine: interesting how well the decades-old Atlas/Roco design has held up; other than replacing the axles due to split gears, my 2 Atlas F-9s are actually the smoothest engines of all of my little fleet. They take very little "juice" to get started, run very quietly, and I think are terrific models all 'round.

I have been running my recently acquired (and 2 -railed) Weaver Geep 38 a lot, and it seems the longer it runs, the better it runs. I am very pleased with it.

Mark in Oregon

PS: There is a very good thread over at "The Railwire" Forum on a P:48 build that someone did with the Red Caboose Geep; well worth checking out...

Thanks for the updates Mark. I forgot to mention the splitting axle gears. I used NWSL wheelsets with gears until I got the P&D brass truck upgrades. Worked great.

Here's a link to that P:48 build. It uses the Des Plaines brass chassis. It's above and beyond what I could do detailing-wise,  especially the grille work and details behind the grilles. Outstanding craftsmanship!   

http://www.therailwire.net/for...ex.php?topic=31130.0

Last edited by Pete M

Well, it's been so long that I had to re-read some of this thread to catch myself up.  

The replacement decoder from SBS4DCC.com finally arrived today.  It's been sitting on a shelf at the post office.

I've loaded up a sound project and away it went.

The moaning sound with this new decoder is the same low level as in the Atlas SDP-35 - that is that it's still there, but easily overcome by the sounds from the decoder.  I ran the fine tuning procedure (CV54 =0 > F1), and the loco has BEMF, excellent slow running performance and nice sounds.

All this has convinced me that the Weaver/P&D motor and trucks/Kelly's adjustable tower; can combine to make an excellent runner with a Loksound decoder.  Sure it still moans a bit, but you can't hear it with the sound off.

I'm happy with that.  Now to add some wipers to the rear windows of the cab and then weather it.

Cheers

Add Reply

Post
×
×
×
×
Link copied to your clipboard.
×
×