Just brought a Biesse Rover, hustle to move it then start to retrofit.

Hi, I'm Sam from Australia. I've previously built a few cnc project machines, all running mach3. I just picked up a Biesse Rover 23 which will be a big step up from the little 8x4 Joes that I'm running now, eventually anyway.
I thought I'd join up here as there are a few Biesse retrofit threads already.

My machine was made in 2000, has a 7.5kw HSD spindle (like most I'd guess), pod and rail bed and yaskawa servos.

I've started stripping it down for transport and am just amazed at the sheer number of I/O on the thing. Coming from a single spindle machine with a G540, about 3 limit switches and spindle speed control, well, it's eye opening.
I've read bevins' thread about converting a 346 and found it pretty informative. He makes the Mesa cards sound really appealing.

Thing is, I will almost certainly never use most of the extra functionality of this machine. I don't run a cabinet shop and have never found myself wishing I had a horizontal drill spindle, let alone a saw.
At this stage I'm planning to remove all the pneumatic attachments from the head. I was initially thinking of just removing the lot down to the Z cariage and then making a new spindle mount but I after removing all the guarding today and taking a look I see that'd mess with all kinds of things and probably be a big step backwards.

Same deal with the pod bed. I absolutely want a flat bed with vacuum. The pneumatic pins are cool but I cut sheets for signs, not cupboard doors so I think they're all destined for the spares shelf. I'll likely just build a sheet bed on top of the sliding rails.
Removing most of the pneumatic functions from the head and the bed looks like it'll cut out about 30 I/0 connections. I was initially just going to cut all the cables (cringe) to speed up the removal (4 days from purchase to lock out from the building) but of course that's a stupid idea. Tomorrow I'll label and disconnect everything from the control cabinet. Normally this would feel pretty daunting but I'm pretty sure most of those cables will get stripped out anyway so it might actually make things easier to follow come rebuild time.

The motion controller, as best as I can tell, is a HSD RTU500. I couldn't find much info on them other than a few comments like 'can only handle 800 lines of code' which won't do. I guess this is where the Mesa boards and LinuxCNC come in. Bevins conversion kept all the functionality of the machine and therefor needed a ton of I/O. I don't think I'll have anywhere near the original I/O and would appreciate any advice on which boards to consider for the servo control, I/O etc.

That said I'd appreciate opinions or advice about my plans for the machine in general. It's a logistics problem at the moment, just getting it moved and stored but once that's sorted in the next few days I'm looking forward to digging into the machine and planning it's new life.

Cheers

Sam

Everyone likes a picture

Sam, welcome to the wonderful world of LinuxCNC and the Mesa ecosystem.
Its great to see another Aussie here. I'm in Brisbane if you happen to be nearby.
I can't help much with Servo based machines but I'm sure Bevins will come along.
I think you will be starting with a Mesa 7i77 which gives you 6 servo outputs and 6 encoder inputs plus some I/O

All i can say is that I think it would be a shame to cripple your machine when you can add 32 of the Mesa 7i84 with 32 inputs and 16 outputs (for a total of 1024 inputs and 512 outputs) to your base 7i77. To me, the extra time taken would be worth it in the long run.

I work with a lot of big flat bed machines. While we don't have any pod and rail machines, We have been considering getting one lately. The two machine types are really built for different purposes. The flatbeds are for sheet goods, and the pod machines are more for working with solid wood parts and glue up pannels or other smaller pre-cut pannels that may need other edge work. If you wanted a flat bed machine you probably should have held out for one, used machines of each type usually are not more expensive than or harder to find than each other.

As to removing features of the machine that you don't necessarily need now, you may come to regret those decisions in the future. Things come up and you end up saying boy wouldn't it be nice if...

While you can just make a flat vacuum table that sits on the pod system, there are a few things you need to be careful of.
1) Table stiffness, you don't want to have any sag or give between your supports.
2) Vacuum supply, flat tables (if milling on a fall-board with no gasketing) need a huge amount more vacuum CFMs than a pod system. Be sure that your vacuum system can supply about 200-300 CFM @ ~15-20 inHg for a 4x8 table (you'll need almost twice that for 5x10.) That usually takes about a 15-20hp vacuum pump.

Things like C-axis and multi-spindle drilling banks and side drilling are still useful on a flat table machine. I know we use our drilling heads a lot (and we are a sign making shop not a cabinet shop also.) It's more work setting all this up, but the end results are excellent.

I got the machine pretty much torn down today. Due to storage limitations I decided to disconnect the control cabinet from the machine which was a bit daunting initially.
I basically removed all the vac pod plumbing today and it’s unlikely it will go back in in future. It was actually in pretty poor shape over all, lots of bits that had broken and never been repaired. Just plugged up and forgotten. Several of the interface plugs from the rail controls had been ripped out and never repaired too. Most of the pneumatic pins have some kind of damage. Not that strange for 19 years of use I guess.

I’ve removed all the rails and end stops, mostly to keep the weight down for the truck that’ll have to pick the thing up.

As I mentioned, all the cables are disconnected from the cabinet and almost all were labeled already so not such a drama. The majority of the connections were for I/O in the base/bed of the machine anyway so will likely not get reinstalled.

One thing that initially had me a bit stumped was the daisy chained cable on the servo drives. I wasn’t sure how this would get interfaced into a 7i77 or the like but, correct me if I’m wrong, I think this interface is just for the original motion control. The 7i77 would connect directly to the drives themselves. Yes?

Anyway, all in all a long dusty day.

hi
yes the servos drivers can connect direct to the 7i77
it semas that the yascavas are chained in CAN system or even directdrive

sgdb can be configureed n more then 10 ways to be moving servos
but they do not have direct access pins to move +-10V servos
you need to remove one cover to see if there is a V24 in

HI, I have a guy willing to give me a Rover 20 if I can get it working again. I am wanting to do something similar to what you are doing Sam. I just want a nesting machine and dont want most of the options the machine can do. I have a very limited budget and would like to just get the machine able to run simple gcode and make the spindle turn on. I know a lot of things have changed and more electronics are available now, and lots of open source controllers. Any advise on if this is worth doing and what the simplest conversion would be, mesa boards or any other suggestions that might be cheaper?? thanks for any advise.
Doug

an old thread, just wanted to note that, afaik, Yaskawa SGDB-xxxN drives only have MECHATROLINK control interface