Weaver - my first attempt.

It's taking a little while for the rollers to be despatched, so nothing to report.

In the meantime, I'm starting a new thread on the rebuild of the Atlas SDP-35.

Cheers

I'm also in a holding pattern with my drive upgrades for now.  

I finally got around to my first tower upgrade but hit a small snag. I had ordered the tower kits without the sprockets and chain as I have lots in stock from all my fiddling about with these drives. However, as it turns out, the Weaver/P&D top sprockets are too large an I/D for Kelly's top shaft. That's because the Weaver/P&D steps up in diameter between the knurled section for the UJ yoke and the knurled section where the sprocket fits. Kelly's shaft, bearing and locknut design is far superior imho, but has constant diameter.

So the key point is you need 1/8" bore top sprockets, not the 3/16" ones that fit standard Weaver/P&D top shafts. You will get the correct sprockets if you order the full kit from Kelly of course. If you need just top sprockets, you can order sprockets from Stock as I have just done, using the part number Kelly was kind enough to send me:

A 6M 7-120804 0.1227 Pitch, 8 Teeth, 1/8" Bore, Plastic roller chain sprocket   

Hopefully they will get here in time for my "6-loco Holiday Upgrade Extravaganza."  

Pete

I'm looking forward to seeing it, Pete.

Meanwhile, I've fitted her up with a Loksound Select L, ready for when the rollers get here.

Should be any day now. 

Cheers

I was sorry to hear about Pete's trouble with the upper gear and have since updated the web site with the following note at the bottom of this page.  

When I was first designing the kit I considered using the stock upper drive shaft from the original locomotive. My decision to replace the shaft was based on several factors.

• I could not source the 8 tooth upper sprocket with a 3/16" bore.
• I do not like C clips.
• The upper 8 tooth sprocket with a 3/16" bore is prone to splitting because the side wall material is so thin.
• C clips will not allow one to adjust the backlash. 

It would have been a lot simpler to use the stock upper shaft but I wanted to provide you with a more elegant solution.

---------------------------- WEB SITE UPDATE ----------------------

A NOTE about Kits without Gears or Chain

This kit includes a new upper drive shaft  that is 1/8" diameter.  The upper drive shaft included in this kit is NOT the same size as the upper drive shaft that came with your locomotive.

The upper 8 tooth gear from your original locomotive will NOT fit the new upper drive shaft included in this kit.  If you are ordering a kit without Gears or Chain you will need an upper gear that has a 1/8" bore. If you are unsure I encourage you to order a kit with Gears and Chain.

---------------------------

Thank you everyone for the support and happy holidays!

Nicely done Kelly!

Thanks to your patience it all makes sense to me now. I had been using Servolink for sprockets and chain. They make a 3/16” bore sprocket but the smallest is 10T which is what I had been using, not 8T. So not only is your top shaft design superior in every way, I will now be able to use SDP 8T x 1/8” and get even a better drive ratio reduction. I had not wanted to bore out 8T 1/8" sprockets to 3/16” because I figured there’s a reason the manufacturers don’t go below 10T on 3/16”. We all know how easily these sprockets can split...  

Your adjustable chain tension, properly parallel top shaft, ball bearings and ability to shim/space the top sprocket into perfect alignment with the bottom sprocket, AND get the UJ spider exactly above the truck pivot, is the only way I know to fix all the issues worth tackling. Thank you for thinking it up and actually getting it into production!

Pete

Yes.  It's a magnificent improvement on the original.

One quick question.  How do I know when I have the tension on the chain correct?

Cheers

Max, chains doesn't need to be tight, mean not tension because each link will find the spoke by itself.

What I recommend is tight manually the tower and return  1/32.(0,7-1mm). 

my 2 cents.

Andre.

Thanks, Andre.

So, push it up until the chain is as tight as it will go and then slide it back 1/32" ?

Cheers

Correct!

~AG~

I agree with Andre's starting point. And then as it's so easy to adjust with this design, you can tweak it until it's at the least noisy  point under load, which ought to be half way between a whiner and a slapper.        

Pete

Thanks, guys.  This isn't related to the above, but when I first installed the Loksound decoder, the motor made that distinctive moaning buzz we all hate.  I changed out the motor for the ball bearing one, and it's still there.  I can get rid of it by switching off the BEMF, but that's just wrong.

Still working on that one.

Cheers

Reading Andre's post made me smile when I remembered a class in engineering school; when the lecturer asked how do we get the correct tension on a stud when we don't have a tension wrench?

Little Johnnie, "Easy.  Tighten her up until the thread strips; then back off half a turn."

OK  Chaps.  The rollers have arrived and the clicking has been traced to when the wheels cross joins in the rails.  I've used insulating joiners, but I've pushed them tightly together, so there shouldn't be an issue.

I haven't started the scenery yet, so once the boards become loaded up they should be deadened.

One thing that has become obvious is that the axle directly below the tower is bent on the P&D replacement truck.  The whole back end of the loco moves up and down.

My main worry now is that the replacement Pittman motor makes that classic moaning sound reminiscent of the old DC motors.  I was hoping that the flash new ball bearing motor wouldn't do it, but it's the same.  The motor on my test board is the old Pittman.  It doesn't make that sound when it's out of the loco.

I can get rid of the sound by switching off the BEMF, but as this loco will eventually be driven by a computer, it needs the BEMF to be on.

The decoder is a Loksound L - the same as in the Atlas; which doesn't make that sound.

A bit disappointing.

Cheers

Max, I think the buzzing could be caused by the PWM frequency. I am only just starting my learning curve on Loksound, so this may be off base but here goes anyway...   

Out of the box, the manual reads like Loksound uses a variable PWM frequency for fine resolution motor control when back EMF is on and motor under some load. This may be why you don't hear it on the test rig with very little load, and why the much lower current-draw Atlas motors don't "drag the PWM frequency down" as far to get smooth start and slow running . It says with some motors this will cause a buzz. It is possible to adjust the PWM frequency to fixed using CV124. Thus should keep it out of audible frequency range and hopefully keep most of the back EMF control.  See page 48 of the Loksound V4 manual. Seems bit 4 can be set to change the PWM frequency.

It may reduce torque or reduce slow running quality a bit but definitely worth a try imho. I haven't tried this yet myself as I only have one Loksound in a Yoder 44-Tonner and that ran great out of the box, no buzzing. But that uses tiny can motors that draw about 1/10th of the current the Pittmans do.      

Interested to see what you find! 

If you are using Lenz try turning off Railcom to see if there is a difference.

Hi Pete and Jim

I'm using a Lokprogrammer, so Railcom is turned off as part of the programming phase.  I have been programming Loksound decoders for about 10 years - always with the Lokprogrammer, so I don't bother with CV's.

I'll have another look at the PWM on the screen.

I've been out this arvo, but now I'll make a video of it running on the rollers on the Lokprogrammer.

I was aghast at how bent ALL of the axles are now I've had a chance to look at it up close.

Back soon.

Understood Max. I started off the hard way with CVs (in Hex even) as I had a Digitrax system in 1994. Next we got the luxury of Decimal, and then the revelation that is JMRI.  A big THANK YOU to the entire JMRI team! 

N.B. I have never used the Lokprogrammer so take this with a pinch of salt...  I downloaded the app and it seems there's a setting under "Decoder > Change decoder settings > Motor Settings > Adaptive regulation frequency. This is a checkbox, and I'm not 100% clear from the manual if it's plain vanilla "off" or "on", or whether it's "off" and "on plus adjustment with the Back EMF sampling period slider underneath". But hopefully a bit of fiddling and user manual interpretation will get you where you need to be.       

Once again I am living vicariously through your efforts in the hope of learning that Loksound 4 XL will be my decoder of choice for Pittman powered locos (where ESU has the right sound set). I am highly impressed that you produce your own sound sets by the way - I can't imagine the learning curve to get to that stage!   

I'm embarrassed to say that I've never used the JMRI Decoder Pro, Pete.  The Lokprogrammer is just too simple.

There are only two options available in Loksound decoders for PWM, viz: 20 kHz and 40 kHz.  Needless to say that I tried both, but to no avail.  Dave Heap has written most of the JMRI stuff for ESU decoders.  He lives here in Oz.  He's very good, but I notice that he replies to LP questions on the forum, so I'm betting that he has one.

I've actually got two - from my old HO days.  One at my work station and one built into the layout.  It saved me handling steam locos more than absolutely necessary

Decoder Pro is very good, but still comes up short against the LP occasionally.  Life's too short for me to bother. 

And, just for Jim I made sure the Railcom was off - even though I was running the loco on the programmer.

Anyway, here is the movie.  It's a bit quiet - I can do it again closer up if you need it.

Notice the way the trucks rock and roll on the rollers. 

No embarrassment necessary Max - I wasn't suggesting you use JMRI. Just that it's about 1,000%  better than CVs in Hex!  

No need to learn it when there's a nice OEM alternative! The Lokprogrammer screens are very nice. At first glance I think the 20 and 40 kHz settings might be for frequency sent to the motor under DC operation. I think the thing you need to change is the "Adaptive regulation frequency" - checked or un-checked. And possibly try adjusting the "Back EMF sampling period" slider underneath. But I think it's the former. Checking and un-checking should change between variable frequency PWM under Back EMF that will drop into audible range at starting and slow speeds, vs. running at a fixed PWM frequency above audible. 

But like I said, I am definitely not 100% sure of this so hopefully Dave Heap can supply the correct answer. I would be grateful to hear his response!

EDIT: OK I just listened to the motor buzz. That sounds like a loco with no decoder!  I don't hear any kind of variable Back EMF or PWM control sounds happening there. None of my DCC decoders of all brands sound like that under any setting. But all my DC only locos sound exactly like that before I put the decoders in.  

WEIRD. Now I am quite confused, sorry.    

Pete  

Ha ha!  That's two of us now, Pete.

I've run this past the dudes on the Loksound forum before, and their stock answer is TURN THE BEMF OFF.

Now, I've done some further testing.  First, I put the Atlas SDP-35 on the LP and ran it with the sound off.  To be fair, there is a tiny moaning sound, but it's easily drowned out by the engine sounds.

With the Weaver, the moaning is audible over the motor sounds.

I've tried the Adaptive Regulation Frequency slider in all of its 5 notches - and with the box checked and unchecked.

You guessed it.  No change.

Because Railroad&Co hates locos with no BEMF, the Weaver looks like being a non-starter.  I might have to box it up and head for O Scale Yard Sale if we can't get a break through.

By the way, what did you think of the dancing trucks?

Cheers