Skip to main content

20-5723-1 Milw. Road BiPolars with PS3

As I doublehead just about all my steamers, it made sense to doublehead two Milw. Rd. Bipolars (they did this in real life, as they did with many electrics). Unfortunately, one runs slightly faster than the other, or one runs slightly slower than the other. I first noticed this when running them independently, one of them had a very slight hitch in its step, forward direction only...I found a slightly loose motor mount screw, tightened it, and am not really sure it solved it or not...that's how slight the hitch or momentary hesitation was/is. Then, I set them on the same line and ran them both at 10 scale mph, and after about 60 feet the faster bipolar pulled ahead by about 24 inches.

Does anyone have a clue as to why this is? I didn't adjust any accel or decel rates.

As for doubleheading these two engines, I guess putting the faster one in the rear would be the best option, as it would be a pusher to an extent, though am wondering if that would wear the motor down prematurely? Putting the faster one in front would generate serious coupler tension over time...had that happen on steamers before...they were some strong couplers.

Original Post

Replies sorted oldest to newest

The electronics are probably the same, so I would attribute the difference in speeds to mechanical friction in the drivetrain. Could be lubrication or friction in the worm gears or spur gears that turn the wheels, or traction tires. I have two Sunset Budd RDC cars and they operate at slightly different speeds running simultaneously on the same track.

MELGAR

Several reasons including Melgar's. So oil and lube everything. Try spinning the flywheel to completely turn the drivers a complete revolution, feeling for any binding if you go inside.

Sometimes it's tach issues like the gap, or flywheels' stripe condition. Some engines are "tight" and need more break in time. Old grease hardening on stored engines.

BTW: That loose screw was a great catch to help the engine last better.

The space created running them at 10mph for 60' seems like it maybe missing a whole stripe or something similar. Even a bind in a side rod causes issues. I've never had a MTH steamer side rod too tight or binding, but I have had others. I have seen MTH RK starter set linkage bent causing weird running.

I run engines that have slight speed variation and usually put the faster in front. That keeps it from shoving a slower engine off the rails in troubled track areas.

If the speed is too far off it may cause issues. That's why we recommend testing them apart before building a consist.

You could test each bipolar with other known correct engines to see if you have a slow or fast one.

From your numbers, the differences in speed are very small. 10 miles-per-hour (full scale) is 14.67 feet-per-second (full scale), which means the model engine is moving over the track at 0.306 feet-per-second (about 3.6 inches-per-second). You ran the engines for a distance of 60 feet, which would take 196.4 seconds - about 3 minutes and 16 seconds - a long time for a difference in relative position of only 2 feet. That would be (2 feet in 196.4 seconds) 0.0102 feet-per-second on the track - equivalent to 0.489 feet-per-second at full scale - which is 0.333 miles-per-hour at full scale. So, if I'm interpreting your numbers correctly, the slower engine is running at 10 scale-miles-per-hour, and the faster one would be at 10.333 scale-miles-per-hour. That is not a large difference in speed between the two engines - just slightly more than 3 percent. That should be a sufficiently small difference in speed between the two engines such that you could run them coupled together.

10 scale-miles-per-hour is a low operating speed, at which small differences in internal friction probably have a larger effect than at higher speeds.

MELGAR

Last edited by MELGAR

Thanks everyone, once again...these are the newer PS3 Bipolars from last year...and in my view they were over-greased at the factory...to the point grease is coming out around the base of the motors  where they attach to the chassis. Is it feasible the slower one has too much grease packed in one or both motors - I kind of doubt this? I did turn the flywheel to try to feel any binding and didn't detect anything. I oiled all wheels and gave each about an hour break-in time running independently.

Very interesting this thread should come up now.  I wonder if it has something to do with the attachment of the tachometer ribbon or the tuning of the tachometers themselves?  For my "just to show I can do it" control system, I decided to use homemade optical rotary encoders instead of those awful tachometers for speed control.

Anthony

Melgar: for my steamers I've noticed larger differences in speed increments in the Triplex (2-8-8-2) vs. the Angus (0-8-8-0). It's so bad I no longer lash them up.

Anthony: I only glanced at the tach ribbons - just to ensure they were seated on the flywheel. That's about the extent of my knowledge.

Another observation among all my PS3 engines, electrics and steamers, is that they will start at different speeds...some don't move until 3 scale MPH, others 2, and a couple need to get up to 4 before they move.

If I had the time, I'd do a side by side video review of the recent MTH and Lionel bipolars.

Do you have an amp meter on your track?.....can you see if one is consuming more amps than the other at the same speed on the same section of track?...that’d give you a clue as to a mechanical or motor difference ,......is the slower one struggling?....if the slower one is consuming more amps, then something is forcing it to work harder.....if not, I’d lash them up with out a train first, faster one in front and see how they do on a short spurt,...if the amps don’t go through roof,...let her eat,...

Pat

@Paul Kallus - My understanding is that 10 mph for one engine is not the same as 10 mph for an engine of a different type/scale in MTH land.  I've seen this first hand and one of the local railroad clubs.  However, speeds between the same engines (same release year) are supposed to be the same...which brings us back to your original issue.  My understanding is that there is also settings in the eprom that determine the speed to engine translations.  Mark D. might have some info on this as he's intimately familiar with the PS3.

Regarding Melgar's comments about friction, etc..., one would assume that with the tach system, MTH implemented some type of PID algorithm to give the "slower" engines more of a boost.  Having said that, if the non tached motor is dragging, that would possibly account for the speed difference as basically the other motor would have to pull the slower motor.  Lot to think about in terms of design issues.

As for lash-ups, I've been wondering (aka plan to ask), if anyone in MTH land has tried starting their lash-up to only discover than one engine doesn't start or one engine starts much later than the others.  Your comments made me think about that again.

A

Test Results:

Slightly faster bipolar: at 16 volts and running at 12 scale mph amperage is between 0.3 and 0.4

Slightly slower bipolar (suspect one): at 16 volts running at 12 scale mph amperage is around 0.7

Thus, it appears something is amiss in the slower one...although after my DCS battery assumptions were wrong, I suppose I need to test a 3rd bipolar to see what it comes in at...because it could be the faster one is not normal...but I doubt it because the slower one continues to exhibit an ever so slight hitch or stutter at slow speeds...barely visible to the naked eye but its there.

FWIW: the suspect bipolar is coming up in conventional upon powering up Z-4000, but then I can address it via DCS remote and "start up."

Should I pull the motors out of the gear box...not an easy job but I can do it if the consensus is to do so...but what should I be checking for? Or, could this be faulty electronics?

@Paul Kallus posted:

Test Results:

Slightly faster bipolar: at 16 volts and running at 12 scale mph amperage is between 0.3 and 0.4

Slightly slower bipolar (suspect one): at 16 volts running at 12 scale mph amperage is around 0.7

Thus, it appears something is amiss in the slower one...although after my DCS battery assumptions were wrong, I suppose I need to test a 3rd bipolar to see what it comes in at...because it could be the faster one is not normal...but I doubt it because the slower one continues to exhibit an ever so slight hitch or stutter at slow speeds...barely visible to the naked eye but its there.

FWIW: the suspect bipolar is coming up in conventional upon powering up Z-4000, but then I can address it via DCS remote and "start up."

Should I pull the motors out of the gear box...not an easy job but I can do it if the consensus is to do so...but what should I be checking for? Or, could this be faulty electronics?

I wouldn’t pull anything apart just yet,....can you run the slower one with the shell off and observe the motors?....get it running as slow as it will dare, and see if you notice anything going odd between the motors,.....if you see one of the motors behaving oddly, turn off the power and check that particular truck for binding, or obstructions. Have you looked at the traction tires closely?....is one slightly slipped off acting as an effective brake shoe?...you’re border line a half an amp difference, I don’t know what MTH calls allowable, but I’d be looking for something binding somewheres,.....me, myself & I don’t like to take anything apart till I have a clear path to go down for diagnostics.......it might take a bit to figure it out, but it sounds like you’ve found the why, now you just need find out who’s responsible,....😉

Pat

Unfortunately, while I can take the shells off disconnecting them and running without them is hairy...too many thin fragile wires connecting lights. I'll mess around some more tomorrow...and report back...I have to say all these gremlins takes time away from enjoying the layout...and uses up time I could be working on my mountain.

Just some notes in passing, about gears---

The motors are  small and of limited torque at the motor shaft.  Their power comes from the high speed at which they can turn.  So a lot of geared=down reduction to reduce shaft speed (rpm) and increase torque.  In postwar items with the larger open frame motors the shaft torque was greater, and the gear reductions less.  With less reduction, spur gears could be seen, sometimes in two reductions through spur gears.  Worm gears were the gift of the Italians (NYC Hudson 700E), with 3 starts on the worm, in postwar typically matched with a helical gear wheel where the the teeth run across the gear face at about 30 degrees to the axle carrying the gear.  This arrangement would not lock up abruptly, but it did have a braking effect.

One important thing to know is that the use of a helical gear driven by a worm shaft is an approximation to a true worm wheel (not to be confused with the track wheel even when mounted on the same axle).  At first, the contact is point contact and the wear rate is high.  Ideally, the helical gear will wear or displace to have a half-moon depression centered on each tooth between the gear sides and maximum at top of tooth.  Generally, this ideal is best accomplished by having a steel worm shaft that is ground and polished, while the worm wheel is of a softer material such as brass {in toys}.  The later postwar worms are have die-formed teeth in softer steel, and eventually these dies became quite worn (I believe this was the later method, from a visual comparison of the 50th Anniversary Lionel 027 UP diesels, to later GP7/9s and O-gauge F3s).  This situation improved markedly before the introduction of GP7s with TMCC (2380s) in 1995--I would guess new dies for the worms may have been obtained about 1990; I lack representative engines from that period as I decided to get parts instead, to replace zinc wheels, plastic gears,  and lack of Magnetraction and 2nd motors .

Unfortunately in 1954, the standard gear tooth pressure angle changed from 14-1/2 degrees to 20 degrees, for metal gears.  (Plans once showed gear teeth, and a 4 on 1 gear face was quicker to draw.)  Unfortunately, plastic gears were kept at 20 degrees, while they often required 25 or even 30 degree pressure angles to avoid tooth breakage.  So my first retrofit resulted in a hot Pullmor, and a hotter truck under it, from running the side gear sets between axles with non-matching geared track wheels.  Moral:  One has to be sure of what one ia looking at when troubleshooting, and it its really hard to see with small mechanisms, although not impossible.

So, on the matter of running in, keep in mind that a worm gear set is really two gear sets, one operating in each direction of travel of the engine,  and each needs to be broken in (forming the little half-moon wear spots).  Obviously, running-in without much lubrication goes faster, but has to be carefully watched.  My advice is to avoid using abrasive paste to help--this ends badly.  It is also immediately to be noticed that end play in the shaft that carries the worm wheel will shift the wear spot; this can be a life-shortening problem if that shaft is also the axle for track wheels.  It can be difficult to add spacers onto the axle, between wheel inside face and truck bearing face.   There is an E-circlet size that can be sprung over a postwar GG-1 axle with this problem. but if a thinner shim is needed, I think on the O-gauge engine on tubular track, a back-to-back as little as 1.07"  between small (like diesel) driving wheels, by pressing the wheels further on, is the better solution.  With the F3s, this keeps the flanges off the switch frog points.  The 1.10" back-to-back seen on so many modern F3s and Geeps comes, I think, from the only comment on this in a Greenberg reprint of 1950's or so service manual pages, for the NW-2 switcher 6220, that dealers make a gauge of 1.10" for it ( an 027 engine (4-digit #, vs 622), 027 track having .03" wider gauge than O track).  Steam drivers are different (larger) and so their flanges are held against the inner rail by the guard rail, where it has large clearance preventing it from holding the smaller diesel wheels off the frog point.  An 1.10" back-to-back is correct for the drivers on steamers; I assume pilot wheels (and possibly trailing truck) wheels have the 1.07" value that standard formulas indicate.

There are standard formulas for the heaviness of grease required in gear sets; application of these formulas to something like crossed helical gears (IIRC), which approximate a worm gear set, have suggested to me a required value of about 600 (sorry, forgot the system of units here, but only about two of the several system are used in sales literature, with fairly different numbers).  Lionel postwar practice (see following paragraph) may have influenced me.  In any case, the concern over this now typical reduction indicates there is a problem with present lubrication.  Possibly the Lion Drive, where only one axle in the motor truck was driven, in diesels, was influenced by this problem.

Postwar Lionel practice appears to have extended the use of these formulas into small mechanisms-- the high numbers that result are due to the high speeds at which such operate in model trains as heavy as O-gauge.  In evidence, I refer to the 1946 626 steamer, which had single motor, but a worm gear on each of the two (end) driving axles, and the follow-on 1948 2333 F-3 NYC & ATSF diesel engines, where the motorized locomotive had 2 horizontal motors, 4 driving axles, and 4 worm gears, again one for each axle.  This would indicate that the fish-oil grease must have had a rating of at least 300 (typical of the upper reach of automotive greases, where a few special purpose greases might get to 350 (bicycle chains and such).  The great advantage of the fish-oil is its adherence, of course, which is superior in that to modern materials (IMHO).  I'm thinking of going with the fish in my limited work, although I'll stick with what comes in factory engines, in case they found something.  (Fifty years ago I did locate a Texaco grease that looked like it could replace the tar often used in the last gear set on bascule bridge spans similar to that on the I-895 beltway (Baltimore MD), which were 155-feet long over trunnions and  about 44-feet wide (two 12-foot lanes plus full outer shoulder, and inner shoulder wide enough for a car).  Those gear teeth were 5" diametrical pitch (that is, on the pitch line, a tooth, valley and small clearance would take up 5 pi inches, making a tooth about 1.57 x 5" thick at the pitch line (about 8" at mid-depth.)

For porous sleeve bearings on axles, I would recommend a 30-weight (SAE weight but don't use automotive engine oil because its additives are harmful) white turbine oil.  This is fairly easy to find; it is used in machinery in contact with food and doesn't leave stains.  One may be able to use 30-weight 3-in-1 oil, but I prefer the other.  These bearings may come in a form already lubricated, but of course the instructions always say "oil."  As I believe such bearings would be outside sourced, they would logically contain turbine oil.

I have a recent Bipolar and will take a closer look at it.  I got it because my father left Belvoir for Japan during the Korean war.  I was 12 and still remember standing on the platform in Baltimore as the train left.  Eventually he reached the ;Milwaukee Road and traveled behind once of these.  He sent back its timetable, which I still have.  But I remember all that because I was never again to spend a full year with all of us at home--48 shots but never got to Japan (well prepared for the CoVid shot with notice 6pm night before--bring it on, I cared not which version and never felt it.  --Frank

Interesting history, Frank...just recently I was reading about Lionel's Italian connection via the "Brute" - the monster Bipolar JLC decided not to put in production, possibly because he thought no one would be able the lift the thing. The helical vs. square cut gears is also interesting...I have a motorcycle with square cut 5th gear that is noisier than comparable helical gears, but I don't mind the noise.

FWIW: I've greased the spur gears on my bipolars with small dabs of white lithium grease...even though the instructions don't state to do this. I assumed they were metal but after reading your post I am not sure. If it wasn't for so many wires I'd feel more comfortable about taking these modern engines apart...I took off the center cab unit last weekend from one of the bipolars because I had overfilled the smoke unit and it took me 3 hours to get it back together - just due to the wiring and trying to get it to fit in without pinching, etc.

Regarding the amperage differences between the two bipolars, I swapped engines from one line to another and determined that it is the transformer and not a difference in the engines. I don't know why one throttle of the Z-4000 would put out more amperage (0.6 to 0.7 amps) vs. the other throttle (0.3 to 0.4 amps) while the engines are running; throttle one is mainline No. 1 and throttle 2 is mainline No. 2 - both are about the same length, number of switches, same number of wire drops. I tightened all connections on the terminal strips, none were that loose, so I don't think that is it.

Regarding the tachometer, it looks okay to me ~ the "reader" is about 1/8" from the flywheel. I took a couple of pictures and noticed a male wire plug that has no female counterpart - its the yellow one in the middle of the picture. Anyone know why this plug exists? It just hangs in the engine compartment.

DSC00641DSC00640

Unless someone has a suggestion, I plan to continue to run both bipolars and lashup them up...perhaps I am only chasing gear variances due to manufacturing.

Attachments

Images (2)
  • DSC00641
  • DSC00640
Last edited by Paul Kallus

I should have mentioned that out the gate Paul, .....Z4000 amp meters are known pathological liars,....I know mine lies to me constantly when compared to my fluke meter.....I’ve seen this amount of variation lots of times,.....if you can’t see any mechanical issues, and the loco runs smooth, then yeah,...perhaps as the other posters mentioned, break that slower in some more to loosen her up, and lash away buddy,....like I said before, if you lash them up and they ain’t fighting one another, then let em eat,...

Pat

Thanks Anthony and John. Am not sure about the # of stripes...I already put it back together...but is it worth counting the # of stripes on the two bipolars - could the factory have installed different #s of stripes on each, would this explain the difference in speeds?

I conducted more tests at high speeds last night. Now, with about 2 hours break in on each, and at 32 scale mph the engines pull apart within a few minutes; after 10-15 minutes the faster one is a good 7-8 feet ahead of the slower one. I am hesitant to lash these two engines together. The faster one would not only be pulling the heavy consist but it would also be pushing the slower one...

@Paul Kallus - No, no, no! I was just curious.  Albeit anything could be possible...would be interesting if the applied ribbon isn't lined up though???  I was curious for my own purposes.  Something you also might want to look at...make sure the bottom of the tach harness is not dragging on the circuit board.  I'll email you a link to some MTH documentation shortly.  I'm leaning more toward gearing, etc...now myself.

Your issue made me think of setting up a gantry for measuring engine speed along a straight run.  Simple enough to set-up but a time eater for sure.

Anthony

@Paul Kallus posted:

I conducted more tests at high speeds last night. Now, with about 2 hours break in on each, and at 32 scale mph the engines pull apart within a few minutes; after 10-15 minutes the faster one is a good 7-8 feet ahead of the slower one. I am hesitant to lash these two engines together. The faster one would not only be pulling the heavy consist but it would also be pushing the slower one...

You're making mountains out of molehills!  At 30 scale MPH for 10 minutes, you have traveled around 550 feet.  You're talking less than 2% speed variation between the two engines!  That's actually very good, and you can't expect it to be a lot better than that.  These two will have no problem being MU'ed.

What he said.

Pick a speed. Run 2 engines separately and note Z4000 amp reading. Amps will bounce around a bit going around curves vs. straights and such. Then couple engines and run at same speed. You don't even have to set up a MU in dcs...just "force them" to couple while running.  Note this is just an experiment.

Now observe the Z4000 Amps. Ideally they would be the same as when separate. But if there is dragging or pushing the Amps will be slightly higher. The question is whether this "tax" is burdensome. For example you could measure the increased current pulling a few cars and make an equivalence...running a MU has an Amps tax of adding 5 box cars or something like that.

Of course if there is some dramatic increase in Amps (like it doubles) when the 2 engines run coupled vs. decoupled, then stop immediately.

Last edited by stan2004

Stan, how you can couple two engines and run them independently...except with two DCS remotes? I can do a test run of a lashup and see how the amperage fluctuates, but I can tell you from experience when one PS3 engine runs faster than another in a lashup...and the longer you run them as such, the tension couple slack can be quite high and will eventually break the couplers. I can get around this by putting the faster engine in the rear and it will push the slower one and pull the consist.

fyi, when operating independently there was no big increases in amps except when smoke units kick on...the bipolar has a intermittent steam generator that kicks on every so often, pretty neat effect. Lionel bipolars have the same feature but difficult to see the smoke.

@Paul Kallus posted:

Stan, how you can couple two engines and run them independently...except with two DCS remotes? I can do a test run of a lashup and see how the amperage fluctuates, but I can tell you from experience when one PS3 engine runs faster than another in a lashup...and the longer you run them as such, the tension couple slack can be quite high and will eventually break the couplers. I can get around this by putting the faster engine in the rear and it will push the slower one and pull the consist.

fyi, when operating independently there was no big increases in amps except when smoke units kick on...the bipolar has a intermittent steam generator that kicks on every so often, pretty neat effect. Lionel bipolars have the same feature but difficult to see the smoke.

I'm not sure I was clear.  I was suggesting a short-term one-time "experiment" to see if the combined current of running 2 engines at the same speed was more, less, or the same when coupled or uncoupled.  So using just one remote you address engine #1 and set to some speed, then quickly address engine #2 and set to same speed.  Depending of if you were initially coupled or uncoupled and you would be the hand-of-God swooping in to mechanically couple-decouple.

But I didn't read that you had the MU setup all ready to go.  So simply run the MU with the two engines.  Obviously you do not need the two engines to be mechanically coupled to run the MU.  They will jointly respond to speed command changes.  So this is really an even easier experiment!  Measure current of the MU with the two engines coupled and then de-coupled a foot apart or whatever.  Do this at a few different speeds.  Do this with the faster one in front "pulling" then behind "pushing".  I realize it's easy for me to sit back in my armchair as you do the work but I'd think this shouldn't take long especially if the MU is already set up in DCS?

Last edited by stan2004

Nothing but problems on these two bipolars...spent another few hours this afternoon tinkering with them...first the DCS system wouldn't find either one, and ended up carrying them to a different block and then it found them...very strange. Then, the slower one stalled out over one Ross Crossover at speeds <13 scale mph...the other one didn't. I knew the panto-track switch was wired backwards...that drove me crazy...then the sounds went crazy in the slower one...sounded like a poltergeist was in the bipolar...the crew talk blasted out at full volume and was distorted, holy scary experience. That, I think, was due to be a loose wiring harness on the one side....thought I was clever in fixing that...only to have the engine stall out again.

I measured pickup roller spacing at 2 3/8" = plastic part of crossover, but why does one stall and the other not? Theorized that even a fraction of an inch in pickup roller spacing difference due to inherent angles of roller, spring tension, can result in both rollers landing on plastic part of crossover at the exact same time. So I bent them backwards a bit and that seems to have helped.

Then, I ran them together, not a lashup yet, independently, with accel rate on the slower one set a bit higher than the other, experimenting if that would help, all seemed well, started working on scenery, and then I hear them in the other room...where are they...should've come around by now...turned out the faster one had a pickup roller forced upward from a Ross switch, improbable but it happened, lost power, stalled, the other ran into it, spinning its wheels...meanwhile I am working away...hearing the electric engines grinding away...you can't make this stuff up, and you can't take your eyes off these trains for a minute. I will just end up lashing them up...life is too short.

Last edited by Paul Kallus

Add Reply

Post
This forum is sponsored by MTH Electric Trains
OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
330-757-3020

www.ogaugerr.com
×
×
×
×
Link copied to your clipboard.
×
×