The Turbines Are Coming! The Turbines Are Coming!

[C855B]

Has anyone else had issues with their Turbine waddling and lunging?  This occurs in both my A and B units.

YES! I am working on this problem at this very moment. I found a bind in the rear truck on my A. I made the very serious mistake of disassembling the truck in search of the problem, only to discover that reassembly is a nightmare. The wheel wipers want to push the axles out of their saddles, causing the other previously-set contacts to pop out. It amounts to spring-contact whack-a-mole.

So far it looks like the bind is, or hopefully was, related to the wheel wipers. They will pop out at the slightest provocation, including tugging on the pickup wire from the frame.

Fortunately my B unit seems to run smoothly at the moment, so I'm going to leave it alone.

As to lack of pulling power with yours... that's got to be related to a bind. Or something. I've been pulling 20 cars up and down a 1% grade even with the separated units for several days with no trouble. ... Just tried it again with the B, no problem, but looking closely while running it slowly I noticed a slight bit of waddle. [...sigh...]

EDIT: I just successfully reassembled the A-unit truck. Bind is mostly gone, "mostly" because I'm still seeing a slight surge at low throttle. I'll try to isolate it further... tomorrow.

[RBrodzinsky]

Has anyone else had issues with their Turbine waddling and lunging?  This occurs in both my A and B units.

I'm also have the most peculiar problem of slow down on grades.  On level track, my set can easily pull 60+ cars.  However climbing a 1.6% grade, on one run I had to cut all but 4, yes four cars!  To make the issue even more bizarre, the set does not spin wheels or completely stop, they simply slows to a perfect crawl.  They'll still actually pull a full train up the grade, just at a deathly slow pace; maybe 1 foot every 10 minutes.  If I disconnect everything beyond 4-6 cars, the set takes off at expected pace.  6-12 is a noticeable slowdown, 12+ is unbearable.   

Mark, have you tried auto-tuning the Back EMF in the decoders? Independently, place each loco on the powered track, with about 6' of clear track, then, with Program on the Main, set CV54=0. Then press F1. The loco will take off at full speed, and then come to a stop. Repeat for the other unit.

[Mark W]

Are using them in DC or DCC?

Should have mentioned, yes, I have the DCC+Sound model. 

[Mark W]

Mark, have you tried auto-tuning the Back EMF in the decoders? Independently, place each loco on the powered track, with about 6' of clear track, then, with Program on the Main, set CV54=0. Then press F1. The loco will take off at full speed, and then come to a stop. Repeat for the other unit.

I haven't had a chance to even re-number them yet, but BEMF was definitely my suspect.  I couldn't find any info on BEMF with Locksound while mobile searching.  Will definitely give this a try! 

[RBrodzinsky]

I haven't had a chance to even re-number them yet, but BEMF was definitely my suspect.  I couldn't find any info on BEMF with Locksound while mobile searching.  Will definitely give this a try!

See section 5.2.4 of the LokSound Select Manual.  Pg 22 & 23. The procedure I describe above is at top of page 23

[reinhardtjh]

I haven't had a chance to even re-number them yet, but BEMF was definitely my suspect.  I couldn't find any info on BEMF with Locksound while mobile searching.  Will definitely give this a try!

The ESU manuals are out of date, but the BEMF section is current.  Go to http://www.esu.eu/en/downloads/instruction-manuals/digital-decoders/ and download the V4.0 manual (Yeah, your GTEL have Selects, but internally the code is all the same)

Check out Chapter 11, "Motor Control" which has mare than you'll want to know about the BEMF CVs and settings.  Particularly Chapter 11.1.3 "Automatic calibration of the motor" on page 48.

[C855B]

Thanks for the link, John. I downloaded the manual for the LokPilot Standard decoders I put in my turbines, and it does not have the BEMF auto-calibration feature. I can turn it off, however, and verify whether that function is behind the surging and "waddling" I experience, or if I'm still dealing with mechanism binds. As to the waddling, I sure as heck hope it's not an axle issue since I really don't want to get into taking those trucks apart again.

However, given this info about the decoder, I agree with the consensus Mark's lack of pulling power is likely related to BEMF values.

[reinhardtjh]

Thanks for the link, John. I downloaded the manual for the LokPilot Standard decoders I put in my turbines, and it does not have the BEMF auto-calibration feature.

Yeah, unfortunately the LokPilot Standard is a way cut down economy version.  The LokPilot V4 works identical to the Loksound V4 and Select decoders minus the sound.  But, of course, it's about $11 (MSRP - $9 street) more in price.

In fact, ESU is working on new firmware for the LokPilot V4 which will mimic some of the "Full Throttle" actions so that you can mix the two on a consist and they will act the same.  The other LokPilot versions - Standard and Nano - do not have enough internal memory for the new code.  Something else to consider when choosing decoders.

[C855B]

I have no regrets using the Standard* since I don't need to mix and match with LokSound versions. I normally run TCS.

But to that end, I did verify that turning off BEMF accomplished nothing, other than it runs like crap without it - low-end control disappears. DC runs better than DCC/no BEMF, even given that DC response doesn't come until 1/3 throttle. It still has a very slight surge and waddle. You can hear the unevenness from the surge in the motor. What was interesting in the testing was when I removed the universal shaft for the rear truck - the uneven running went away. That's why I took on disassembling the truck. It's improved, but not gone.

Here's more... I have (now had) a virgin DC turbine sitting on the shelf while I was debugging the decoder situation. I just broke the seal and ran both units separately. A-yup... surging and waddling. Not major, but enough to notice if you look hard, and you also hear the same unevenness in the motor as the set I've been tinkering with.

For myself, I've determined it's the nature of the beast which will hopefully diminish with break-in. Also, I noticed last night that the gear tower is a lot drier than your typical Atlas or Kato, where you at least find some lubrication. It may be a simple case of too little or even no lube. While I still have the first A-unit apart I might try a drop or two of gear oil and see where that leads.

EDIT: * - Turns out I have LokPilot V4.0 decoders. I ordered Standards and that's what was on the invoice, but what I received was four V4.0s.

[AKNscale]

Mark, have you tried auto-tuning the Back EMF in the decoders? Independently, place each loco on the powered track, with about 6' of clear track, then, with Program on the Main, set CV54=0. Then press F1. The loco will take off at full speed, and then come to a stop. Repeat for the other unit.

If you're going to do that I would recommend doing both together, it will match them together nicely. If you go to sbs4dcc.com he demonstrates this in a video on an F7? A and B pair. I believe one unit was a LOKsound Select Micro and one was actually a LOKpilot too.

[CBQ Fan]

I got mine on the layout tonight. Very very impressive.  I am glad I pulled the trigger on these!

[Spades]

Noise and passenger complaints aside, turbines couldn't do passenger service. They had no steam generators, for one thing. ain is probably the closest, and, sadly, with no known pictures I can't even prove that.

They didn't use the turbine's bleed air to heat the water?

[C855B]

Last night, I isolated a bind in the A-unit I had decoderized and then had to remove a truck - the chassis lid was pressing against a flywheel. The chassis is similar in arrangement to the early Con-Cor/Kato PA and U50, split horizontally with a "tub" holding the motor and mechanism, and a lid of chassis metal keeping everything in place. Apparently there is very little clearance in there and I think my primary sin on reassembly was cranking down too hard on the screws. In the end I had to leave one loose.

This is going to be a trouble area for a lot of folks. A couple of times on reassembly I seated the screws all the way down like I would on anything else, but in these cases it was evidently so hard the motor would not turn. I checked whether the motor popped out of its seat slightly, but found nothing. If I find the screws come loose down the road, shims or washers may have to be added to add clearance.

Frankly, considering the infuriating wheel wipers, the gear tower clips that come loose on the slightest tug, wires that break off in handling (three so far, including one cold joint) and the bizarre situation with the chassis-induced binding, they have ignored decades of experience provided by Kato, Atlas, Bachmann, FVM, Athearn and pretty much any modern N scale locomotive designer ('cept BLI), and created, in my estimation, a virtually unmaintainable mess. Clearly they're beautiful models, and they run OK when new, but IMO they run a high long-term risk of becoming shelf queens or even bench queens.

[C855B]

They didn't use the turbine's bleed air to heat the water?

There was a hot water system to heat the tender to keep the fuel from congealing. I think it was just hot water, not steam for auxiliary use. I only know this from the recent tender rebuild program which removed the hot water piping and corresponding insulated jacket.

[peteski]

Frankly, considering the infuriating wheel wipers, the gear tower clips that come loose on the slightest tug, wires that break off in handling (three so far, including one cold joint) and the bizarre situation with the chassis-induced binding, they have ignored decades of experience provided by Kato, Atlas, Bachmann, FVM, Athearn and pretty much any modern N scale locomotive designer ('cept BLI), and created, in my estimation, a virtually unmaintainable mess. Clearly they're beautiful models, and they run OK when new, but IMO they run a high long-term risk of becoming shelf queens or even bench queens.

Unfortunately it seems that the Chinese designers aren't taking any clues from proven-to-work-well existing designs.  Funny thing is that back in the '90s those Chinese designed realized that Kato split-frame low-friction-trucks was really good and copied it on many models.  They abandoned the wheel wipers and wires, and chose to use the Kato low-friction/wire-less design. Nowadays they seem to have reverted to the old-school N scale mechanisms (with all the modern electronics and detail parts putting a lipstick on a proverbial pig).  They came around full circle, back to the old design. 

It makes no sense to me why they use wheel wipers and wires instead of what Kato and Atlas do.  The parts count is higher and labor cost is higher (the wipers and wires must be manufactured and assembled/soldered).  But yet, those designs are getting approved by the mothership in US.    If it was me I would send some sample N scale models to the Chinese (Kato locos for mechanisms and overall shell engineering, and MTL wooden reefer for a reference on how to model the gaps between the wooden boards) and tell the Chinese designers to learn from those examples.