RV4S Driver Belts (and OpenPNP)

[dogboy]

Hello all,

I want to replace the motor drive belts on my RV4S.
As many of you know these are marked up as T2.5-950

My question really is where do I buy good replacements from?

Searching the net brings up many suppliers from RS to eBay with various prices up to £30+ each and and different types of materials etc.

My existing belts measure exactly 12.6 mm wide, so I assume that's what I should replace them with.

What is the ultimate drive belt for these machines?

My belts don't look too bad, but I am 'upgrading' my RV4S to run with OpenPNP and a Duet3 6XD controller and really I would like to put new belts on as I have the robot assembly ripped apart at the moment and it would be good to know they are new without any issues.

If anyone is interested in this, I could start another post documenting my progress, which may or may not be successful!

Thanks
Paul

[trev]

I would be really interested to see your progress with this. Be interesting to see what repeatable placement accuracy you get.

[dogboy]

I just priced up two timing belts (T2.5 950 13mm) from Bearing Shop UK. Including VAT and postage (£2.40) it comes to a total of £84.02. I nearly fell of my chair. LOL.

Can't seem to find any other alternatives apart from some Dunlop branded belts, but these are described as budget belts and the only other make I found was Optibelts but they only had 12mm width (totallybelts.com).

Any advice on what are the best replacement belts to buy would be helpful. I can always order the belts from Bearing Shop but just thought I would ask again, before coughing up 84 quid.

Thanks
Paul

[trev]

Don't waste your money until you have proven the machine will do what you hope. Am sure you will have it apart a few times as your project develops.

I spent money on stuff thinking it would be a good investment only to find a project takes a different path and you end up wasting money.

[Mike]

If they look OK I'd leave them for now - don't think I've ever seen anyone reporting belt issues.

TBH I'd question whether an OpenPNP port is worth the effort - the inverse kinematics for the RV are very non-trivial, e.g. there are some motion paths that need to be split into two moves due to the limits of the geometry.

Then you have the issue that the downward camera has an angular offset from the head which varies with arm position, and of course part rotation also varies with position.
The original developers needed a machine grid calibration to get enough accuracy.
And even if you get all that sorted the feeders are a bit crap.

An interesting project if you have time to burn, but you easily could end up with something that isn't quite accurate enough to be useful for anything.