Need to amplify 0-5V signal to 0-24V (supply is +-12V)

I'ld go with #2.Get an opto to handle the soleniod current and put it in series with the soleniod between the + and - 12 (so it connects the soleniod across 24 v).

I don't know why this topic posted twice, once with the entire body as the subject... I did not do that.

Anyway, thanks for the response. While waiting for responses I came up with a different solution. I made a non-inverting amplifier and an inverting amplifier out of some op-amps. The gain of each is 2.4 which turns the 0-5V into 0 to +12V and 0 to -12V, referenced to ground. I used the output of each amplifier as one lead to the solenoid. It works! Sort of. The coil requires 60mA at 24V and the op amps quit increasing voltage at about 42mA. So my coil gets 0 to 18V for now. If I parallel another op amp in each amplifier circuit I think it will work. This is not for a finished and doesn't need to be perfect. I'm in R&D and just need to control a solenoid valve with a microprocessor to see if it will work. The EE's in Japan will sort eventually out the circuit details.

-Alex

What you want is a switch, right? A solenoid is either on or off. So, use an optocoupler to turn on/off a transistor whose emitter is at -12V and collector goes to the solenoid, which goes to +12V.

You might want to use a schmitt trigger to get good snap-action and keep your switches out of the linear region.

Good Luck! Rich

Actually it's an analog solenoid. I need to have an analog voltage from 0 to 24V at the solenoid.

You're not going to have enough headroom at +/-12V supply for the 0-24V output across the solenoid-in any simple way. A first cut, in simplified form, could look something like this-assuming the solenoid is really "analog": View in a fixed-width font such as Courier.

. . +12 . | . |< . +12 .-------| . |\\| | |\\ . IN >---------|+\\ |/ | . | >---| | . .--|-/ |> | . | |/| | | . | -12 [Re] | . | | | . | | | . | | | . .-[R]--+--[1.4R]---|---------+---------> (+) . | | | . --- gnd|-+ [Rs] . gnd | | . .-----------|---------+ . | | | . | | [Rs] . | | | . | [Re] +---------> (-) . | |\\ | | . '--|-\\ |< | . | >---| | . .--|+/ |\\ | . | |/ | |/ . --- '-------| . gnd |>

. | . -12 . . . .

You can get 23.-something volts across the 400 ohm solenoid with a single-ended transistor output stage, but have to do a little trick to resolve the difference between the opamp being referenced to 0v and the solenoid voltage being referenced to a supply rail. +--------------------------------+--- +12v | | | R1,56k R2,12k | +--/\\/\\---+---/\\/\\---+ | | | | | __ | | + +--|- \\ | ) |OP1>---+ )Lsolenoid 0v----|+_/ | ) ?? (1H) | | R3,12k | / +---/\\/\\---+ \\Rsolenoid | / 400 R4,12k | R5,56k | - Vin----/\\/\\---+---/\\/\\---------------+ | __ | +--|- \\ R6,100 |/e |OP2>------/\\/\\---|pnp, 360mW max 0v---------------|+_/ |\\ dissipation. | ---+-- -12v

V(across solenoid) = (+12v) + Vin*R5/R4 - (+12v)*R2*R5/R1*R3.

If R2*R5 = R1*R3, then V(solenoid) = Vin*R5/R4.

OP2 must have rail-rail output, (ie CMOS) and the max V(solenoid) will always be one Vbe less than 24V.

--
Tony Williams.

Silly me..... you can get the same result with one biassed opamp.

+--------------------------------+--- +12v | | | | | + ) )Lsolenoid ) ?? (1H) | / \\Rsolenoid / 400 R4,12k R5,56k | - Vin----/\\/\\---+---/\\/\\---------------+ | __ | +--|- \\ R6,100 |/e |OP2>------/\\/\\---|pnp, 360mW max 0v---+ +--|+_/ |\\ dissipation. | | | +--/\\/\\--+--/\\/\\--+12v ---+-- -12v 12k 56k

LTspice files 5to24.asc and 5to24.bmp posted to abse.

Vin swings from 0.5V to 4.5V with 10mS slopes and the +12v rail has been given a 100Hz 2Vp-p ripple.

--
Tony Williams.

Nice circuits but 1) the OP is gone (the 30 second attention span was exceeded by the minimum SED response time) and 2) most "analog" solenoids with which I am familiar are *current* programmed, voltage control is not used without some form of either displacement or magnetic flux feedback.

Thanks for not mentioning the circular scenic route to get to that simple result Fred.

You hinted along those lines in an earlier post, but I was still hoping he had some sort of positional feedback in there.

--
Tony Williams.