returning to a RPC for a new old robot Yaskawa UP6 with XRC control

Good afternoon all. I have a new to me Motoman UP6 robot arm with a XRC control. I revived my old home made RPC to run this machine. The RPC ran my old mill with no problems. i switched to a VFD when i got another machine. I had both machines running off the one VFD (one at a time with a 3pole double throw switch).
Well one salt water flood later, the VFD was toast, but the RPC was above the high water mark. so some new wiring and we are back in business.
Now for the new 20 year old robot. The Motoman XRC control is set up for multi voltage, it has a big transformer in the enclosure with 3 taps for 240, 480, 560 in, and one 240v, or 208v output,not sure what the output is actually suppoosed to be just yet. Motoman said+ 5%, minus 10% on voltage is ok
I hooked it all up, and i was able to boot up the old robot, but ran into trouble while changing the CMOS battery. I thought i killed it dead. but a few phone calls and it is back up and running. But one of my problems at least i think, I had a low voltage leg. I traced this back to the transformer still being hooked up, and running 208 into the transformer and getting 175v on one leg out of the transformer. I figured I did not need a transformer at all, going fromm 208 to 208. So just bypassed the transformer all together. and i have
L1-L2= 207
L1-L3= 238
L2-L3= 214

l1-gnd= 118
l2-gnd= 172
l3-gnd = 119
ground here being the case of the control

I am afraid to run that high voltage manufactured leg into my old control. I only get one shot at this thing. burning up a control board in the 20 year old robot control will pretty much brick the whole operation

I was reading a voltage control thread gman4405's . i like the idea of moving the manufactured leg to some system that is not so voltage critical. But i dont' know which system that would be. I near as i can tell. there are 3 ish subsystems . 1. the main control (computer), 2. the servo drives, 3. the brakes and other . and then all the low voltage stuff for the encoders, and maybe the brakes as well. Since there is no dumb spindle motor, i can't go there. The whole load on the system is tiny 1.5KVA. I am running a 3hp home made RPC
So the question is How can i reduce the manufactured leg down a bit in voltage. I don't know how to get a reading on this while under load. I have not even got one servo to move yet, let alone all 6 servos together.
is there a calculation for a capacitor that uses HP of the RPC

then my next question is pretty basic, But i am running a 3 wire 220vac single phase system , hot, hot neutral. I think i want to put in a real ground for the robot. But in my junction box the ground and the Neutral are still jumpered together, I know that is not the modern way. but that is how it is set up at least for now

in a modern system the neutral stays isolated until it get to the first box at the service disconnect, correct ?

If you have an RPC, you would expect the output, in an ideal world, to be identical to a high-leg delta from the power company.

That gets you 240V between phases, with two legs at 120V to neutral/ground and one at 208V.

You are actually a bit low on everything related to L2, which is what is expected with an RPC.

Do you have any wiring diagrams? Generally you would look to connect anything not requiring three phase power across the original supply, which is L1-L3 in your case.

It's likely this equipment doesn't require a neutral at all.

thanks for the response. i might be making this a issue on my own, but there is some stuff that is just not rite here. I finally took a closer look at the specs. The machine wants 240vac 3phase, +10% -15%. so i am in spec on the system, and as "someone" said i am ok on the high end and close to out of spec on the low side. But take a look at the picture, and i am out of spec on the voltage to ground stuff.
As to putting the manufactured leg to non critical system. I do have schematics. but i can't see that any of the individual pieces require 3 phase power. I will have to look closer. Again This is a 6DOf robot arm, so only servos, no spindle motors at all. the biggest servo i think is just under 1200 watts. if anything, maybe the the servo drive system which is a all in one sort of thing, is what takes the 3 phase.

I have to admit i am a bit under educated about how the RPC does its thing. I originally thought that the manufactured leg would be lower than the power company's 2 legs, but read that the manufactured leg is higher, so i went with that.
today I have a 121Vac at L1-N and L2-N.
and now i can't understand the R-gnd, S-gnd and T-gnd issue from the picture. This is just what i have been calling L1,2,3. Is this where i am having a issue because i don't have a dedicated Neutral and ground?

thanks for your help so far

its like they intended the machine to be be corner grounded delta.

i don't know why it would matter. any sensible person doesn't run a connection from phase to ground. its run from phase to phase or phase to neutral.

then again, its possible this machine was intended for 120/240v single phase.

check the internals, nothing would supprise me.

btw, if your peak power consumption is less than half of the drive's peak rating, then it won't matter. all modern servo drives are basically a VFD on steroids. they all just convert the ac into dc, then back again into 3 phase to drive servos. . 3 phase produces cleaner dc power but the drive already has the bandwidth to compensate, and the single phase derating is simply for the input rectifier and the capacitors.

you mention the transform has taps. for 3 phase you would need to change a minimum of 2 taps for each voltage, more likely 3 taps.

there are some older systems which used 12 SCRs to drive 3 phase motors directly from 3 phase ac. i doubt anything built in the last 30 years uses this method.

i am reviving this thread. I am trying to get back into this project. I still have a similar problem. BUT i have switched to a VFD . Now i get nice equal voltages between the phases. But i also am still getting voltage to ground on all the phases as well. This is at the main disconnect for the robot with the knife switch off. and before the fuses inside the cabinet
last night i got

L1 TO L2 = 228
L1 TO L3 = 228
L2 TO L3 = 228

L1 TO GND = 115
L2 TO GND =122
L3 TO GND =126

if i remember correctly the folks at motoman told me not to run the robot on a VFD. And i am not planning on energizing this circuit. But i wanted to test it this way to see if there was some kind of major problem with my RPC. This is a home made RPC that is now 15 years old and was just barely above the high water line during a salt water flood 4 years ago.
My take away from this is that I kind of need professional help. Unfortunately i have not met any local electricians who even know what a phase converter is. I did meet a high power electrical engineer who insisted that i could generate 3 phase electricity with a transformer.
Any one have any ideas here? I have a older system where the ground and neutral bars are connected in the main breaker box.
Could there be significant change in these readings if the circuit is energized??
Where? Should?? 2 of the legs get together so as to put 240v to the ground as the manuals wiring picture shows to be correct, and then not have voltage on the 3rd leg to ground