Three phase and IBM

On 19/03/2011 11:49, Jochen Kunz wrote: I beg to differ. Y or D and voltage are two separate issues, and you probably need to get both right. Three-phase motors seldom bring out the star point even if they are Y connected, and will run from a Y or D supply of the right voltage impartially. Other stuff (basically, electronics) may be connected between two phases, or it may be connected between a phase and neutral. If the latter, it must by Y; otherwise it doesn't matter. The voltage must still be right either way. The electronics could even be connected between a phase and a "neutral" other than the star point (see below). So beware! Finally, I have come across kit (not a computer, I admit) that ran off 415V three phase, with the controls supplied by a little 415/110V transformer. In the UK, voltage is seldom an issue - almost all three-phase stuff is 400V or 415V, whether it's Y or D. Continental Europe goes in for 690V as well, and this is starting to appear over here for largish motors, which can be nasty. But I wouldn't expect to find that on anything as old or small as a 3880. In the US, all sorts of strange voltages and connections were used for three-phase supplies. Even the internet probably doesn't have a complete list! 208V was common, usually Y (so that phase to neutral would be 120V). I think 240V Delta was used, with a centre tap on one of the windings providing a neutral, so you got your 120V from either of the two phases on that winding. (I don't know whether it was normal to ground the neutral, but be aware that it was NOT the star point!) ISTR seeing 277V somewhere as well. This would be 160V phase to earth - I have no idea what, if anything, uses that! This means, inter alia, even a five-core mains cable could be a delta - three phases, centre tap of one winding, and ground - with no star point. My advice to you is find all the rating plates you can on the equipment, both outside and under the cover panels, and see if they agree on voltage and phasing. Rating plates on individual motors may help. Once you have done this, and wired up the appropriate supply, blip the power on very briefly to make sure the motors spin the right way. If they don't, swap two phases around. (Choose two phases symmetrically placed about the neutral if the neutral is not the star point.) Then, and only then, do a smoke test. I am not sure about equipment of that date, much less IBM equipment of that date, but I'd guess that undervoltage is less likely to be harmful than overvoltage. (Some switch mode power supplies may draw too much current if the voltage is low, and kill themselves that way.) So err on the side of lower voltage. HTH, Philip.