Mitsubishi FA20SA: Using B-axis as a Turning Spindle

Hello!

Recently my company bought an optional rotating axis for our Mitsu FA20SA machine. The "authorized" service team already had problems with installing it (the axis rotates 10 times more than it was told), but they told us little-to-nothing about how to properly use it. We figured it out mostly, but we have prolems finding a generator suited for "turning" cylindrical parts (mostly ejectors with some sort of shape on it, for plastic molding).
Have any of you had success with this kind of machining, and willing to share? We use 0,25mm (0,01") brass wire, but any other parameters are welcome too, for comparison or otherwise.

Thanks for the answers!
Balint

Hi Balint,

It's several years old, but everything still applies and should get you going. Read up on "Turn and Burn" in EDM Today's archive section.

213_Spring_Feature_EDM_Today

Bud

Hi Bud!
Thanks for the info, it is very helpful, truly. I make sure that we will try out the written method soon. Still, the main problem is, that if the surface that we want to cut - or rather "turn while burn" - is uneven, even by a smallest (0,03-0,1mm in radial direction, according to the dial indicator) amount, the wire breaks. We can't find the right parameters for the generator. Even is we can cut the shape we want, neighter the size of it, nor the surface won't be as we desired. We use the E-pack that came built in with the machine, we tried all that we could think of, parameters to very small, very thick, variable land machining generators, with and without powermaster, but nothing seems to work. If i cut something on a 10mm rod (while it rotates with 30-50 rpm) the powermaster shows that it cuts waaaaaay aboe 100mm.

If we cut parallel with the turning axis, it's mostly goes smoothly, but if we cut within an angle with the axis (like cutting a cne) the smootness if gone. This is also true for the skimming cuts. Not to mention, if we neet to cut in parallel with the radius towards the center... that if an absolute nightmare. If we can cut in the same direction, but from the inside out, it is way better.

I dont know (heck, i even doubt it) that there is an universal e-pack for it, like there is for the standard mahining, but i can't believe, that it should be this painful. i'm not even sure what rpm we need to use exactly, i can only say, that high rpm durning the first )rough) cut leaves spiral bumps on the surface, that can't be removed later - or at least we could not.

P.s.: Sorry, if i composed the original thread badly, and also please excuse me for my (presumably bad) grammar. If anyone has anythink like this info, please share it, we certainly need it.

There are many different epacks in the machine, try different ones. Also wire speed and flushing can make a difference. Your speed is going to have to go way down in order to get a decent finish...like.01 ipm and under.

I hope it's not a 3r b axis, those things are shit.

It is... An integrated "B" axis, bhich is parallel with the X axis. (WTF?) The guys, who installed it made some mistake, and we have a 10 to 1 factor between the programmed angle (or rpm) and the real angle/rpm. if i proram "B100.0" iot will turn 1000 degrees. Insane.
Also, we can't really adjust down the feed that low, because the "adaptive control" will owerride it. We tried to tinker with both the wire speed and with the fushning, but it seems like that the spark is not strong enough to burn down the material fast enough.

But, we never actually trued to rough the part as it was shown in the article that Bud provided. We will try out the thing that way, if it will work like that, i will be very happy (my boss won't bother me all day with the stuff)

Thanks for the advice anyway!

veiszb2 said:

It is... An integrated "B" axis, bhich is parallel with the X axis. (WTF?) The guys, who installed it made some mistake, and we have a 10 to 1 factor between the programmed angle (or rpm) and the real angle/rpm. if i proram "B100.0" iot will turn 1000 degrees. Insane.
Also, we can't really adjust down the feed that low, because the "adaptive control" will owerride it. We tried to tinker with both the wire speed and with the fushning, but it seems like that the spark is not strong enough to burn down the material fast enough.

But, we never actually trued to rough the part as it was shown in the article that Bud provided. We will try out the thing that way, if it will work like that, i will be very happy (my boss won't bother me all day with the stuff)

Thanks for the advice anyway!

Click to expand...

A lot is in the roughing op. This will remove most of the material allowing the rotating passes to be skim passes. For the finish passes I would turn adaptive control off as well as pm. Set your feed rate and adjust manually until you find a sweet spot. You will probably need multiple skim passes as well. I made small parts for patella implants, the roughing took 30 or so minutes the finish took up to an hour to finish. These parts were only 1.5" max od and .35 deep.

Hi Bud:
As always, that article you wrote and referenced was a highly worthwhile read...thanks very much for the link.
To the OP: as others have pointed out, you need to be allowed to experiment without the Boss-Man breathing down your neck demanding to know why it's not done yet, in order to be successful doing this kind of work.

One of the things I've found is that CNC machinists in general and wire guys in particular have become much more "By the Numbers" kinds of workers than was the case "way back when " when I was just a kid entering the profession.
This changed culture discourages experimentation in place of the formulaic solution...we look to a chart, or we look to a tooling catalogue, or we look to a set of calculations to solve our problems, and for the most part it works well.

This is not one of those cases; you need to find a way to trick the machine, and the best way to do that is to understand why it's not working.

To that end, I encourage you to re-read the section where Bud writes about the theoretical aspects of how the turn-and-burn process differs from the cutting of a simple profile in a flat slab.

Be aware particularly of the fact that when you cut a rotating mass, you are effectively cutting at a much higher feedrate than the nominal traverse rate of the wire across the job.

Another way to look at is that you're cutting an effectively much thicker part than is implied by the depth of the contact zone and as Bud states, you have far less contact area, so you cannot put as much power into the wire.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Thanks for all the help you guys provided, we were able to figure out a method that works for us.

We have tried to "rough" the profile multiple times, rotating the part by a certain angle after each profile (like a polygon), but it was slower that the straight cut with the continously rotating part. We also had some major pain with the cutting of the very rough profile that this method was made. But in the end it was successful, also very slow.

With the direct profile cutting with continous rotation (by a certain rpm) we were also achieved some progress, but after the cutting, the part still had some (very strange and rough) part of a spiral shaped remaining material, it was very annoying.

So we tried the middle ground. Now we cut the profile in one step, but the part rotation is according to the length of the path. This way we have to rotate the part by 57 150 degrees per inch, (actually just 5715, since our axis rotates 10 times more that instructed) or 2250°/mm. It is similar to the theoritycal thread cutting. This way we can rough the part relatively fast, while using the factory generator sets. We only need to mess around with the corrections. When there is some remaining material, the machine will rotate back the part, not trying to advance forward, and cut it properly. We can also leave both the PM and the adaptive control on, and we end up with a smooth surface.

I know that this method sounds bizarre, i also had doubts, but it actually does work for us. Even my boss was impressed (a little).

Thanks again for all of the help. We really made progress thanks to your advices.

Another question: is there a way, that we can remove, or increase the 60mm/min feed limit of the PM and adaptive ctrl? Through the roughing phase the machine could go faster, but this limitation holds it back. The feedrate also effects the B-axis, this way the rotation won't be faster than 60 (or 600 in our case) degrees per minute. This os the only downside (so far) of our new method.

Thanks,
Bálint

Hi, with reference to your x10 problem on your B axis, I also installed a B axis here in the UK to a Mitsubishi machine, and there is a way of making it x1, what control do you have on your machine? Is it a touch screen machine? Also which B axis do you have Hirschmann, ITS?

OK, from your other posts I can see you have an advance control, do you know how to access the hidden service pages? Need you to look at a servo parameter for me SV_018 PIT and tell me the value in the B axis column.

Hello,
We have an ITS B-axis, but i don't know the part number right now. To be honest, the 10x movement is not a problem right now, since the method i described earlier works thanks to this. But i guess knowing the right settings would not hurt. I know how to access the hidden settingy, i will look into it as soon as i can.
Thanks!

Hi,

So, we have an ITS RSI-42 B-axis. I didn't founund what you have asked, i haven't seen such like. This i what i see: Imgur: The magic of the Internet
Please, tell me if i have looked the wrong screen or something. Also sorry for the img quality, for some reason the machine does not like to take screenshots anymore.

The service guys messed up with installing this thing, they could not figure out the parameters. they tried multiple parameter sets, and the axis was inoperable for a month and a half. When they finally made it working, they were so happy about it, that it was left like this, moving 10x more that instructed.

Yes, this is the correct place, on the right hand side of the screen you should see "page down", press this once to see the next set of parameters. I need to know what SV018-PIT is set to?

Okay, it's 360.

Okay, you could try changing it to 3600, see what happens then. Remember to switch off the machine and then turn it back on again for any changes to take effect. Also I think you need for ready to be off before it will change.

Thank you! I will try it when i have the opportunity.

haha, classic M+E rotary axes install.
I feel your pain