Axle Hung Motor Gearbox (AHMG)

[IanT]

Engineering in Minature Doddy

IanT

[Doddy]

I take it it is the November Issue?

http://www.teepublishing.co.uk/eim/issues/engineering-in-miniature-november-2015/

OR

http://www.teepublishing.co.uk/eim/issues/engineering-in-miniature-october-2015/

??

[IanT]

Part 1 of the article appeared in the October issue and generally described the design & layout of the 5" loco bogie and why a brushless motor had been chosen to power it (small size, high power & better reliability) and which (100W) motors were actually used. It's an interesting read but perhaps there is nothing so different to existing practice that it would warrant any remarks from a G3 perspective.

Part 2 is in the November issue and describes the theory of how a brushless motor works and what is required to make it usable in practice. Jon then goes on to describe an electronic building block approach to building a BL motor controller and how he used available 'modules' to build his own controller. He's used an ARM Cortex card (in a Arduino format) as part of the mix and written the C++ code to drive all four motors and provide full driver controls.

Interestingly (at least to me) he's very clear that he needed the extra compute power of the ARM to do this (a fast interrupt routine watches the encoders on each motor (there are 4 axles) and drives the motor phase changes.

He also describes the use of PCB layout s/w to design a 'finished' controller PCB (after a vero board prototype was used to prove the design worked). The final PCB design (using plated through holes - PTH) was then commercially produced for him by a third party company.

So whilst there wasn't so much new information in the first part of this series, there was a great deal of new (to me) information in the second part. Well worth a read if you are at all interested in this new motor technology.

Regards,

IanT

[Gavin_B]

An update on the Brushless motor. 

I knocked together an ugly proof of concept that included half a brick to see if the brushless motors with no gearbox could do any useful work.  (it was pointed out that it was not really useful work by Tim's better half)

2 motors pushed a train of 10 wagons at slow speed, as seen here.

https://youtu.be/wXZXCIT3-Ok

still some work needed as to speed up the motors but as a proof of concept am happy.   Total cost,  £14 for two motors, arduino uno £9, triple halfbridge £12,   Brick stolen from the rockery

[IanT]

Interesting update Gavin & a good video too - thank you.

Regards,

IanT

[IanT]

As you may have gathered - this is not a high priority for me at the moment but I have had another look at it recently.

I'd arrived at the first design shown and it has lots of issues and I was beginning to feel I'd spent too much time trying shoehorn the HQ7P motor in. However, I needed a rest from my current work and so had another look at things and the various problems with my (then) current design. To keep Ralph happy, I also used Delrin gears (0.5 & 0.8 MOD) from Muffett.

I think this new design is viable but it has a few issues and I'd certainly want to build & run one before recommending it. The gear ratio is 10:1 which is hopefully sufficient for this application. I've used commercial gears & bearing components without modification and assumed the use of laser-cut frames. Some further work would be required (soldering motor clamps) and possibly a couple of simple turned bushings to limit axle movements, although I've used the gear flanges for this purpose currently. Neither drawings are complete but they are sufficient for my purposes (to test clearances etc) - "thinking about it" drawings if you will.

I probably will make a 'test' gearbox eventually but I have other projects that need to take precedence at the moment. If anyone wants to have a go though, please let me know and I'll provide dimensioned drawings of the key areas (mainly the frame plates) and provide details of the components used. I've not costed this unit yet, as I don't have Muffett gear prices to hand.

Regards,

IanT
   

[Gavin_B]

Thats looking good Ian.

I changed the software for the brushless motor controler and now can manage slow speeds and a respectable running speed.   Did not manage to get any video before running into problems with wheels turning on the axles.

Those motors have so much touque they will just refuse to stall.

They are proved enough for me.

Gavin

[IanT]

Sounds very good Gavin - watching your progress with great interest.

Regards,

IanT

[classicdelights]

Hello Ian

Did you receive my message regarding drawings etc. for this, I sent it off-line?  I will get the frame parts laser cut and put a couple together, we will then know how much it all costs.

Regards

Phil Flint

[IanT]

Hi Phil,

I've been away but have just replied to your PM.

Regards,

Ian

[Peaky 556]

I will get the frame parts laser cut and put a couple together, we will then know how much it all costs.

Phil,
Thanks for taking this up, I will be most interested in the outcome and expect to buy or build several for myself.
When it comes to laser cutting though, my experience has been that unit costs are very low, but the setting up cost is very high. This will make it uneconomic to make small batches. Maybe you are adding it to another (bigger) order?
Once a prototype is built, and taking advantage of the low-cost Muffet gears and economies of scale with laser cutting, I am pretty confident we will have a cheap but effective gearbox for these motors. Hopefully also versatile for many traction models. The market quantity might even approach the number of motors that John sold to us, which was something like 200 wasn't it?
Good luck ?
Tim

[classicdelights]

Hello Tim

You are right about order quantities but initially I always use Model Engineers Laser.  Malcolm High provides a very good service for small numbers of items and the costs are reasonable but clearly not as cheap as getting 200 done.  To my mind that is the only sensible way to go, at least while the design is tried out.

Regards

Phil

[MikeWilliams]

When it comes to laser cutting though, my experience has been that unit costs are very low, but the setting up cost is very high.

I am very surprised at that.  I've used four different laser cutters over the years (Bingley, Coventry, Bletchley and Leighton Buzzard) and nine has ever charged a setup fee.  If you do the artwork and send them a dxf file, then the setting up by them is minimal - really just adding the too path.  But I do agree that if the order is very small its hardly worth their while and larger firms will have a minimum order charge, but that's not really a setup charge.

Mike

[Peaky 556]

Mike
I probably misunderstood the reason for the charge structure at the local laser cutter in Derby (Accrofab), but the bottom line was a minimum charge of £85 (or thereabouts), on top of which is material cost and possibly some other factors. In my case the order total amounted to the region of £130, which was substantially less than I'd been quoted by three other outfits for the full complement of bits I wanted.  Interesting to know where to go for small quantities...

By "artwork" do you mean anything more than just a 2-D CAD file?

Regards, Tim

[MikeWilliams]

Sorry Tim, I didn't write that very well.  I did indeed mean 2D CAD saved as a dxf.

Mike