DCC Conversions produce coffee grinders - help

[learmoia]

So.. Something I've always wondered about the Bearden Method..  I've never tried it.. but with removing the inner bearing and block and adjusting the hex head to hold the worm in place (which seems like a trial and error PITA)...   Couldn't you remove the inner bearing but keep the inner block to hold the worm in position, then you don't have to be exact with the hex head?

~Ian

[u18b]

I'm not sure I follow you.

The "bearing"  and the "bearing block" are the same thing in modern locos.

Now, if we were talking about the Atlas/Kato RS-3 from the early 80's..... well that one had a bearing like a BB in a plastic case.  So theoretically you could remove the bearing and leave the case.

But modern locos (ATlas/Kato etc) use a solid plastic bearing.

For example, here is a Kato RS-2 worm set.
http://www.katousa.com/images/926530.jpg

The bearings are plastic.

By the way, notice that it has a true u-joint, so this loco would not need to be Beardenized.

[mmagliaro]

Here's the results of my first experiment.     I had a funny moment while I was test-running the resulting engine, when it suddenly started emitting an enormously loud whirring buzzing sound.  I thought, "Oh no!  I've ruined it, Charlie Brown!"

... then I realized the engine was running just fine, and it was some guy next door running a chain saw.   You can't make this up.

Now to the experiment.  I did this to a Kato F7, which is not even one of the engines in Ron's recommended list in his paper.
It does not have a hex-nut coupling from the worm shaft to the flywheel.  Instead, it has a round cup with 4 little spider fingers.
I was a little concerned that removing the inner bearings would allow the whole worm shaft to flop around, but no.  It worked.

First, here's our stock victim.  Trucks out, running on the bench.  I ran it for 5 minutes in each direction to make sure that it was fully warmed up and there would be no fooling around due to things not being seated, not having been run in a while, etc.


As you can see in the next photo, at 7 volts, it draws 240 ma.


Then I opened the frames, pulled the worms out and just put them back in, the way someone might do if they were
servicing the engine.  Sure enough, I could not make it run as well after I screwed the frames back together.  The current was up in
the 300's, and all the futzing around with the bearing blocks, loosening the frame screws, and whatnot could not make it
run at 240 anymore.  I COULD make it run quiet, but not as efficiently.

Next, I removed those inner bearing blocks.  Here's a pic of the guts.


After doing this I tested it, and I tweaked the position of the flywheel coupling on the end of the shaft as Ron recommends, so I got just a wee bit of play in the worm, but not so much that it rubbed at either end.  Sure enough, it ran at 220 mA.

I thought, "There ya go!  About 10%" 
So yes, it certainly did work.  I can't say it ran any quieter than I was able to get it before.  But it definitely drew less current.
The ammeter never lies.

But then, I thought, "What does Victor know?  I'm going to drill out those inner bearing blocks, and put them back in, even though
he told me not to."  It turns out that Victor knows a lot.
With the drilled-out bearing blocks, I got lousy current readings in the 300s again.

Not willing to give up, however, I persisted, because this simply did not agree with my physical understanding of the engine.
If the inner bearing holes are drilled out so much that the shaft never touches them, then it should be like they aren't there, except
that they act as a stop to keep the worm from sliding laterally back and forth too much, which ought to be a good thing.
So I removed them, checked, and drilled them out a little bigger.  THEN... before I put them back in, I polished the faces of them with 1000 grit
sandpaper, because it seemed to me that when they were back in there, the worm just would not move back and forth AT ALL, which is
not right.  It needs to have a wee bit of space.

So.. with the drilled-out, adjusted inner bearing blocks back in...

BWA HA HA HA HA HA!



That's right one seven three..... 173 mA!   More than 30% lower current consumption.

From 240 down to 173.  And bear in mind, I had the headlight LED connected the whole time, so the percentage decrease is actually
even better.

Now... to the track.  I cannot honestly say that it runs any quieter than some other Kato F7's of the same vintage I have.
Sometimes I think it does.  Sometimes, I think it's my imagination.  But once again the meter does not lie.
Pitted against another F7, this modified one will start and run 0.5 to 0.75 volts lower than the other one (about 2.5 vs 3.25).

The two can still run well together (plain DC, no speed matching here),
but I think I really have to modify the other one now.  CURSE YOU, RON BEARDEN!   

[peteski]

Interesting result Max.  I was thinking that removing the inner blocks also gets rid of inner thrust bearings, but since there is no load on the worms (trucks are removed), the thrust bearings aren't really engaging.

To be honest, I'm really surprised by the difference in current consumption. It seems like way too much of a difference on an unloaded drive train (again, no trucks).  I guess I'll have to someday try this to believe it myself.

[mmagliaro]

Interesting result Max.  I was thinking that removing the inner blocks also gets rid of inner thrust bearings, but since there is no load on the worms (trucks are removed), the thrust bearings aren't really engaging.

To be honest, I'm really surprised by the difference in current consumption. It seems like way too much of a difference on an unloaded drive train (again, no trucks).  I guess I'll have to someday try this to believe it myself.

I am just as surprised.  That's the reason I tested it with no trucks.  I wanted no other forces involved other than the drive shafts and bearings.
Pretty shocking that there is that much power waste in the drive shaft itself.

[u18b]

Max

As far as inner bearing block friction.....  I'm amazed at a 30% reduction.

But as far as noise goes.... I'm not surprised.  You inadvertently violated one of Ron's qualifications.

That spider thing is a floating u-joint.
That will allow the worm shaft to flop around a tiny bit.

Ron's Rule says that this procedure works on rigid shafts.
There must only be one moving point... and that is at the outboard bearing.

I don't think any Kato locos truly qualify.
Atlas and Life Like have lots of candidates.

But the current drop is amazing.  Even better than I would have thought.

[u18b]

The reason your results weren't worse, is because the distance from the spider to the shaft is so short.

So the flywheel held the shaft "fairly" straight.

But Beardenizing is a disaster on longer locos, like 6-axle.
This...


Now THAT thing would flop like crazy.


Here is a better example.

Atlas GP15.
Single rigid shaft.
Hex nut into and properly adjusted into the flywheel.
Whisper quiet.

This was one time that I experimented with two thrust washers (for some reason), but the process works better with one or none.

[peteski]

I am just as surprised.  That's the reason I tested it with no trucks.  I wanted no other forces involved other than the drive shafts and bearings.
Pretty shocking that there is that much power waste in the drive shaft itself.

The results were so unusual that it might be worth it to repeat the test.  It is almost like something else (besides removing the bearing blocks) has affected the result. Put all the parts back, then test the current comsumption, then "Beardenize" it, and check the current again.

[Power Stroke]

I still contend that this should not be an issue at all, and while performing Ron's mods, of which I have on a few locos, (sometimes working,sometimes not) This is only addressing a symptom, and not the root cause(s).
One of my thoughts is that the insulators are at issue, as well as torque on the frame screws. I have also found that the frames are tweaked (inconsistent clearance).
Regardless if you paid 20.00 or 150.00 for one of these locos, they should not turn into a buzzing, grinding mess after you open them up one time to install a board or service them, but what do I know.

[victor miranda]

may seem a little like drift...

Max,  7 volts and .175 amps is what I expect from a loco pulling a train.

it is spinning fast at that draw?
I kinda am thinking you ran the voltage up to 7 volts?

mmmm  1.22 watts --> 7 volts and .175
      1.68 watts --> 7 volts and .240

I am thinking pre-loaded thrust caused friction.
just not sure how it happens. or was reduced.

Properly done, the inner block removal can and often does reduce noise.

I want to know how to get the loco back to quiet with the blocks...
Read that as I have not figured it out.

I am curious about the short Kato double joint
when it is over the trucks and hauling a few cars.

I was thinking it may allow enough off center to have the worms wobble.

victor

[mmagliaro]

To address all your concerns, and believe me, they are identical to the ones I had:

1. I had this engine apart, back to original, then with bearing blocks, then without them, quite a few times.
So to address Peteski's concern, yes the behavior was consistent and repeatable.

2. Ron, yep, I kind of figured that this engine did not really apply to your fix, because the flywheel attachment isn't rigid,
and I was afraid that the spider would rattle or let the worm wobble.

but surprisingly....
3. ... it didn't.  And this connects to Victor's concern.  I have run this engine on the track with a 20-car train.  No question,
it starts at significantly lower voltage than another of the same model that I have, and it will easily "catch up" to that other one if I place them both on the track.    I believe Ron is right, that the spider is so short that I am "getting away with it."   What it DOES mean is that
the spider is probably so short that there isn't enough U-joint action between there and the inner bearing to allow for smooth operation if there is any minor misalignment.

I am going to do this to the second F7, in the name of science, and see the behavior is repeatable.  This time, I will put little Sharpie
dots on the bearings to 100% guarantee that in my reassembly attempt, I have them back in the same exact orientation
as when I remove them.

I really learned something.  When I saw the spider, I expected this to totally not work. 

[peteski]

Max,
I wonder if you could measure the current draw of the motor running by itself at 7V?

I ran a one of my Kato motors (not from F7, but it was a standard 5-pole straight-wound motor) without flywheels and it consumed about 92mA at 7V. Its static resistance was around 40 ohms.

[mmagliaro]

Ugh... Good idea, Peteski, but truthfully?  I don't want to take these apart again.
I have just returned from doing this with a second F7.  Unfortunately, I could not attempt this on the second engine because....

The bearings and the spider coupling were completely different!   Yikes.  I had no idea there was this much variation in these things.
The second engine uses the much superior (in my opinion) bearings with the floating ball-shaped bronze centers.  The first
F7 had plain plastic bearings with no bronze in them.  The coupling to the flywheel is more like a cone with four little nubs on it,
rather than a round insert.  So without the support of that inner bearing, this thing definitely would flop all over.

It draws about 155 mA, just as-is (20 better than my other one after that one was improved!)
EDIT: And yes, this is the engine that starts and runs at a HIGHER voltage than my modified one, even though its
bench current at 7v is LOWER than my modified one
This doesn't surprise me.  Those round-ball bearing are really nice.  They just turn and float until they find a happy place with minimum
friction.

So I'm afraid I don't have a second example.  I have some bare Kato motors, but I am not sure they are the same as the one in the F7, even
though they look the same.

Oh!  And while we're at it, what's with the fast gearing in these things?   My older F3's were geared really slow, and I love them.
These are definitely faster.  I seem to recall that F3 and GP50 trucks have a slightly different gearing that is slower than the F7.

[jpwisc]

Wow,
The data that flows in these threads never ceases to amaze me. I believe it is all the best in one place.

Back to the original question: I know it was mentioned at least once already, if the noise level changed during a decoder install, and presuming that the bearing blocks were reinstalled correctly, try loosening the frame screws slightly. I have over tightened screws before and had similar symptoms. Once the screws were loosened a little: no problem. Try the simple things first.

[mmagliaro]

And still the data flows...

So, I took the F7 where I had drilled out the inner bearings, and I replaced them with bearings from my scrap box that had the ball-shaped bronze centers.   As I suspected, the current draw dropped still further, from 173 to 154, which curiously, is the same
as the other F7 I did not modify (which already had those bronze ball bearings).

Thinking about it, those bearings really are ideal.  You get a smooth, bronze bearing surface, but in a ball that can rotate to adapt to any slight misalignment in the driveline. This is much better than the plain plastic blocks that were in there.  With those, you pretty much have to hope the hole drilled through them will end up in a perfect line with the shaft.

And the running quality definitely improved.  The engine is quieter and the starting voltage dropped even more, to about 1.75v.

Now... what do you all know about the lineage of those round ball type bearing blocks?  Are they older or newer than the simple plain plastic block type, and if they are older, why did Kato (or anyone else) stop using them?   Cost?   They seem like such a beautiful bit of engineering.