View Full Version : Belts & pulleys & advice please

25-04-2012, 06:50 PM
Hi guys

I would appreciate a bit of help please.
We are trying to source items needed for a profile cutter conversion. (1.5m x 3m).
Similar to the one on this page: http://www.donedeal.co.uk/for-sale/tools/3218065
This is a slow moving machine.

Various factors are pointing us towards using belts and pulleys to drive the two axis.

My first stumbling block is:
Since this beast has a very heavy (but smooth) gantry, I doubt that a direct drive on the pulley would be best suited, so my reasoning is to go for
- at least a 1:3 reduction pulley on the drive (is this enough?)
- source a bigger oz/sq motor for torque at low speed (thinking of 12NM nema 34)
- use as small possible pulley to drive

Second stumbling block:
Looking at availability of pulleys/motors/belts I have realised that I have no experience in how pulleys are mounted on the motor. I know you can drill them out to the desired motor shaft size, but how is it secured properly?

Thanks in advance

Web Goblin
25-04-2012, 07:12 PM
Hi, having worked on many of these old machines in the past a word of warning. Most of these would not go any faster than 3M/min and we primarily designed and built for gas cutting only. We had retrofitted some of them for plasma cutting but the top speed remained the same. This cantilever type machine only has a drive system on the middle rail and uses the outside rail and bearings as a sort of counterbalance. The problem with this machine moving too fast is that it will whip on the outer end of the cross beam. Acceleration should be slow and smooth to prevent this. As you say these machines are quite heavy but were driven by small motors. The picture is not too clear but does it have gearboxes fitted to it?
What are you thinking of converting the machine to do?


25-04-2012, 07:39 PM
Hi Ian

Thanks for the quick response.

Correct, the machine is SLOW and will remain so. The aim is to replace the current faulty tracing system on it.
The current motors on it is a small 19v DC pittman motor with a huge reduction ratio. Most likely more than 200:1. Replacing the faulty motor is around $700 before shipping, taxes and freight. For that price I can go a long way towards CNC.
The machine will not be doing anything else but what it has been doing. Cutting very slowly through thick metal.


Web Goblin
25-04-2012, 08:09 PM
Those optical tracing systems were not brilliant but ok, if I remember correctly it was a Linatrol HL90 but it was a long time ago :distant:. The main problem with tracing is the amount of storage space you use up when you want to keep the image for later use. The CMC 2.5 didnt have a great memory. As you say parts for this machine if still available will be expensive, I think ESAB bought out BOC a long time ago, and they arent cheap.
As for fitting pulleys you can easily drill them out on a lathe and simply drill and tap it to take a grub screw and use it to clamp it to the shaft, or you can drill it in position on the shaft and hammer in a roll or c pin to secure it.
Another option is to use a taper lock clamp. These need the pulley drilling out to a much larger size but allow for easy fitting adjustment and removal of the pulley.


25-04-2012, 08:26 PM
Thanks a lot Ian

What were your thoughts on the motor and reduction selection? Any suggestions on size and power needed on a belt drive?


Web Goblin
25-04-2012, 08:41 PM
I wouldnt like to guess on that one. We had a similar machine which I think was powered by 280w groschopp motors through gearboxes. Off hand I cant remember the reduction for the gearbox but it wasnt huge. Incidently I am using NEMA 34 motors on my machine. These would probably have enough power to run the machine as long as the bearings are in good condition and it is setup correctly. Not sure about the reduction you would need for a belt drive.
How are you going to get the CMC 2.8 to run the steppers? Does the machine have seperate encoders fitted to it for positioning?


25-04-2012, 08:57 PM
Nope, I am way to ignorant to integrate it with CMC, planning on replacing the whole HL-90 with Mach3. It means the tracing functionality is gone, but opens up new possibilities and as you said, more storage for templates and GCode for popular jobs. Also, the tracing is not great when it comes to narrow lines, for instance tracing a gasket. It works, but sometimes the tracing will work against you. With CNC you have better control over each line cut.
It means that now we will have to create CAD first which will slow turnaround on new items, but we might save some time on others.
Will eventually build a new table, but for now just looking to get it up and running again.


Robin Hewitt
26-04-2012, 08:58 PM
Nowt wrong with an optical head so long as you have someone there to kick it when it reaches that awkward corner :beer:

I'm just drawing up a 2.5m x 1.25m plasma table and Gary has promised the 2.7m profile rail will arrive unbent.

I plan to use roller chain. I know it's supposed to be a big no-no because of the cogging but the price of the 04B 6mm chain has plummeted, it doesn't stretch and with a 20t sprocket my drawing says the cogging error will be around 0.076mm, about 3 thou.

3:1 down on to the sprocket gives me 40mm travel per motor rev which divides conveniently down to 0.1mm per half step and seems optimal for both rapids and slow cutting.

You only need large motors for a heavy gantry if you need large accelerations. If you don't need large accelerations to avoid a plasma torch blowing the corners off on thin sheet then NEMA23's have a much better speed range than NEMA34's.

If it is free rolling, and you only have the inertia to contend with, then the maths is easy. If the gantry weighs 100 lbs then you know that a force of 100 lbf will accelerate it at 1G because that is what you would happen if you dropped it. If you driving force was half the weight then you would have 0.5G and so on.

To calculate the driving force you need to know how far the motor will move the gantry in one revolution then divide that in to 2 pi.

eg: If I fitted a 0.5Nm motor to move my gantry 4cm/rev the driving force is 0.5 * 6.28/0.04 = 78.5N = 17.6 lbf

Meaning, if I can keep the gantry weight below 17.6 lbs I might hope for 1G acceleration from a 0.5Nm motor.

Play with the numbers, see what you can come up with.

26-04-2012, 09:25 PM
What feedrate do you currently get/need on this machine? Knowing that and the mass of the gantry along with what Robin said you'll be able to work back to find an appropriate motor.

I'm just drawing up a 2.5m x 1.25m plasma table and Gary has promised the 2.7m profile rail will arrive unbent.

Any length profile rail will arrive bent. It's a matter of by how much!

27-04-2012, 04:28 AM
take a look at Lenze. they provide belts, pulleys, clutches & brakes.


Web Goblin
27-04-2012, 06:16 AM
some good advice from Robin for you.
I had been looking for an old BOC Falcon manual that I used to have but cant find it anymore, must have been recycled. This might have given me the original gearbox specs. Removing the CMC 2.5 and the HL90 might be your best option in the long run. The Burny is a good little controller for a low cost machine, pity about the lack of graphics though. You needed a Burny V or 10 for graphics. It understands G code, Word address and ESSI files. Nice package in a small box.
If you down the PC route as you say once you have drawn up your parts they are easier to store for future use. One benefit you will have is if your customers can supply you with a ready made cad file, this would just need checking then off through the post processor and your ready to cut.


29-04-2012, 07:26 PM
AH crap! I wrote a reply on Friday, never posted it.

Robin, thank you very much for taking the time to explain the calcs involved. It is very handy to know.

Ok, so I can calculate what minimum size motor I need, but is there really a problem with going to BEEEG (12Nm) motors? I am just sold on the idea of doing a conversion on this machine, but migrating everything over to a new table at a later stage. If I get them boys big enough, I can do what I want at a later stage. Ofc I will be matching them with the right drive and power. I am sure Gary will help me with that ;)


Robin Hewitt
29-04-2012, 10:31 PM
is there really a problem with going to BEEEG (12Nm) motors?

Just looked at one... Holding torque 12Nm, working torque about 8Nm which it can hold up to about 5 revs/sec before starting to fade.

That power was obtained with 6.2A of pure 110VAC which could be tricky to replicate.

Your real problem is that the hobby market has standardised on 2 phase 200 step/rev motors and now you can't buy anything else. It's good for a NEMA17 but not for a NEMA34, the inverse square law is against you.

Thinking that a monster motor will solve all your problems is a mistake, been there, done that. Fitting a smaller motor seems so counter intuitive nobody will believe it could possibly work.

If you want power maybe servo motors are the way to go, but I have no experience of them.