Need design for pos/neg voltage source

I need to control a bath circulator/chiller. The analog input control accepts 10mV per degree C. Positive voltages result in temperatures above zero C, negative voltages (referenced to analog ground) result in temperatures below zero C. I've build positive voltage supplies, and negative voltage supplies, but no supplies that swing both ways. What's the best way to do this? I was thinking of using a positive supply but instead of connecting the ground directly to the analog ground, connecting to a voltage divider with a 10:1 split. Would this work?

Reply to
lektric.dan
Loading thread data ...

If you don't have to generate the supply for the input control (i.e. if you're just getting a voltage from somewhere else), _and_ you can stand a hit to your common-mode rejection, you don't have to have a negative supply. You just need a differential amplifier on your front end that'll shift the voltage up to whatever you want the zero reference to be, and won't violate your op-amp's common mode input requirements while it's doing it.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html
Reply to
Tim Wescott

l
s

Dang I hate getting old. I've done something like this a few (ok, about 8) years ago. I used an instrumentation amp to create an offset voltage for an accellerometer used to meastre tilt. But I'll be switched if I can remember how I did it. Don't have my design notes from that far back. Could you please explain in simple words of one sylable or less (grin).

Reply to
Dan Major

But, what's your input coming from? Do you want a knob? What range does it cover? And, what is the input impedance you have to drive? What power sources are available?

Probably you want to start with a regulated supply, and float it (don't connect either the (+) or (-) terminal to ground). Then, one resistor divider connects to the ground reference, and another, adjustable divider, connects to the analog input.

Reply to
whit3rd

Yes, maybe, depending on what the input control impedance looks like.

Something like this:

.----------. +-----------------------------------+----------|+5 | | | | | | | | | | .-. | | | | | | | | | |249R | | | 2K00 200R '-' | Isolated | .-. ___ ___ | | | | |

Reply to
Spehro Pefhany

I have read the other responses and now i know i have a different take on this.

Some questions: What is the temperature range of the fluid bath apparatus? What is the maximum current required at the control input? Have you considered normal DC amplifiers? Do you need to provide more advanced control than setpoint? Do you need to provide an error signal instead of a setpoint voltage? Do you want to have controlled temperature vs. time profiles? What is the temperature drive capability of the bath? What is the heat load capacity of the bath? etc., .

Reply to
JosephKK

I'm sorry. I was checking for responses for a while, didn't see any and gave up. Now here, a month later, I find that some people *did* peply and I missed them. Again, please forgive my rudeness!

What I have is a Thermco Scientific Neslab RTE circulating bath chiller. It has provisions for a remote control of temperature. The voltage input for the setpoint is 10 mV per degree C. We need to chill down to about -10 degrees C (machine limit is +200 degrees C to

-20 or -25 degreed C). This will provide circulating fluid to chill a seres of condensers. The ovens that will work in conjunction with the chillers will operate at up to 800 degrees C. Although the main controls on the chiller DO provide for temperature profiling or ramping, we will operate at a single temp setpoint. Considering the operating temp of the ovens and the mass of the reactors inside them, plus the mass of the condensers, I think we'll be able to chill down the condensers a lot quicker than the ovens heat up. The chiller also has an output of 10mV/degree C of actual temp. I plan on using some inexpensive LCD panel meters to monitor the voltages. I have a fairly well regulated 12v DC supply that is powering other instruments/stuff on the panel. I appreciate any comments or suggestions you can give.

Reply to
Dan Major

Thanks. That looked a terrible mess until I cut/pasted it into notepad and used the right font. I don't know what the input impedance is, I assume it's high using such small values for control voltages. I don't know if I can easily find these resistor values, but I'll play around with what I can get, keeping close to the ratios you mention.

Reply to
Dan Major

Stated.

May we assume moderate (10K to 1M)?

Is this still relevant?

Apparently not.

Seems to be command rather than error.

Does not seem to be applicable

Presumably sufficient.

Presumably sufficient.

? Maybe more to discover that is relevant.

Not a biggie. Usenet can be like that, or someone can be trying to catch up on a few weeks of posts for any of plenty of reasons.

Off to reread root post if i still can.

Reply to
JosephKK

Now with better context from later posts, i recommend a simple potentiometer with a simple op amp follower. With +5 and - 5 volt supplies and if you want to do it nice voltage references for +2 V and

-2 V.

Do you have LTspice or similar? It is free and i can then post a simple circuit for you.

Reply to
JosephKK

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.