Suitable power supply for these servos

[lsteele]

Hello,

I have three electro craft e240 servos with these specs (I bought them on ebay - these were the supplied specifications):

DC Permanent magnet Servo Motor
0.22 Nm Stall Torque
1.7 Nm Peak Torque
3.1 Amps Stall Current
24.5 Peak Current
4700 RPM Max Speed
48 VDC Max Voltage
1.1 kg Weight
4.5 kg Radial Load
2.3 kg Axial Load
7.8 v/kRPM Voltage Const

And I'm driving each one with a uhu servo controller. I've built three uhus, and for testing purposes I've been using a pc power supply - 12V, 10A. My router is a MDF construction with a working area around 24x36 inches.

My question is: What kind of power supply would allow me to get the most out of these servos? I know it says 24.5A peak, but I presume this is way in excess of what they'll actually need in practice. Could I for instance use 4 pc power supplies in series? That would give me 48V, 10A (for all three motors). Obviously that's way short of the theoretical max current consumption for three, but would it be adequate in practice?

I really have no experience in this area, so any advice is appreciated. Incidentally cost is something of an issue, but I'm quite happy to build stuff (as opposed to buying things of the shelf).

Any suggestions are greatly appreciated!

Thanks,
Luke

[lsteele]

Incidentally, I'm not looking for someone to present me with a finished power supply schematic for my setup. What I need help with is determining how much current I need to be able to supply these with (@ 48V)...

[H.O]

Hi,
Since you have three motors and each motor can pull 3.1A continous I'd go for a supply that can source atleast 10A.

You may be able to use 4 PC powersupplies in series but I'd say a simple unregulated linear supply would be a better choice since it has no problems supplying much more than rated for current during short periods (acceleration etc).

A 36V/10A transformer, bridge rectifier and a capacitor, 22000uF or there about would do it nicely. It won't hurt with a tranformer capable of more current.

/Henrik.

[lsteele]

That's great - thanks.

One more question - Since the uhu's need 12v for the logic side what's the best solution here? I would presume a transformer with an extra 12V winding, right? Or could I use a 12V regulator? Incidentally, you mention 36V - the motor specs say they're rated to 48V. Are you suggesting that 36v would be adequate?

Thanks again,

Luke

[H.O]

Hi,
The transformer secondary voltage is 36V AC, when you rectify and filter that you'll get around 50V DC.

A separate supply for the logic is probably the easiest way in this case since bringing down 50V to 15V is going to be a bit complicated. If your transformer has an extra 12-15V winding you could build a small regulator circuit to produce the 12VDC needed but a small switchmode powersupply or a wallwart is probably the fastest/easiest way in this case. Make sure it's 12V DC and not AC though.

/Henrik.

[Al_The_Man]

Why not the PC 12v from a HD connector?
Al.

[H.O]

That's a good idea except for the fact that you loose the isolation between the PC and the drive (provided there is any to begin with).

/Henrik.

[Al_The_Man]

That's a good idea except for the fact that you loose the isolation between the PC and the drive (provided there is any to begin with).

/Henrik.

That I realize, Incidentally I do not use isolation on any of my systems, the type/make of drive is not indicated.
Al.

[lsteele]

Thanks for all the suggestions. I'm going to spend some time finding a suitable transformer in the next couple of days.

Incidentally all this has prompted me to ask another question which has been perplexing me for a little while. A lot of the break-out-board designs I've looked at (I've opted for this one for the time being: http://pminmo.com/PMinMOwiki/index.p...s_Interface_ST) have an external power supply @ 5v with the ground apparently connected to the ground on the parallel port, i.e. the computer's ground. If my understanding is correct I'm very unclear why this should work. Surely there's no guarantee that the computer's and the external power supplies grounds are the same. Isn't it possible that there would be a potential difference between these two grounds and so a current could flow, which I presume would be a bad thing and not was intended at all? Or am I completely misunderstanding what's going on?

What prompts me to ask this is if I have one power supply feeding the uhu logic and the bob which has its ground connected to the pc's ground, will I then be losing isolation?

Anyway I tried hooking up the above BOB with an external 12v supply and it works fine, so I guess my concerns are unwarranted, but I'd like to know why!

Thanks again.

[Al_The_Man]

If you read past post on this subject, there are quite a few.
It is the difference between practicing complete isolation between systems or setting up a common ground plate, where all systems are made common to this would normally include the service ground, PC power common, power supplies, all shields etc.
It is mostly a case of personal preference, some manufacturers will state their product is made for total isolation, and I think it is done in many cases to avoid users implementing the Common-Ground system and doing it ineffectively and causing problems.
Personally I prefer this system and have used it effectively with PC based systems in Industrial environment's for many years.
Al.