help for linear control of current

If the coil is 120ohms you'll need +/- 120V to get 1A. Say the supplies are 130VDC.

If the coil is 9 ohms, at 1A you'll have 121W+ dissipation in a linear driver at 1A.

Assuming from the NI stuff that this is more-or-less a one-off, and if this is acceptable, you could use a power op-amp and put the coil n the feedback to get constant current.. Apex has some that might work. but pay attention to the SOA curves- they're not especially cheap, and a failure might do nasty things to your coils if they are not fused.

Or build the equivalent out of discrete parts, but you'll be lucky not to have a learning opportunity and/or sound effect or two on the way.

Best regards, Spehro Pefhany

If it's really a one-off, and really slow, maybe a programmable lab / "system" power supply? Sounds like he may already be using labview.

nt

me

ch

As Spehro points out an amp through 120 ohms is a lot of power.... 9 ohms is a bit more manageable. If you need precise control of the current then you don=92t want to use the coil R as sense resistor. (it changes as the coil warms up.) The =91classic=92 way is an opamp, pass element and sense resistor.

V supply |\ | Vin----|+\ |/ | >--rr---| +-|-/ |>

| |/ | | coil | | +--------------+ | R sense | GND

A relay could change the current direction... but you don=92t want to be switching things with the current on. Arcy sparkies and such.

George H.

If the coils are "coils"... i.e. they have inductance, why wouldn't you use a switcher and avoid all the dissipation? ...Jim Thompson

If he wants to drive current in both directions then slapping one into the feedback loop of a power op-amp will work, but at the cost of needing

To get by with one 0-120V (well, 0-140V really, because the amplifiers from Apex have significant overhead voltage) you'd need two power amps, in an H-bridge configuration, with current sense feeding back and all that raz-matazz.

I think it would be better with the intended load on the top side, the collector. To protect the transistor, you use a zener from the C to grd rated at max the collector can handle. This way, you can use a variety of source voltages.

This is just my opinion.

Jamie