Is there a way to convert 0-10VDC to 4-20mA using resistors only? I am building a test stand that has four 0-10VDC input signals and four 4-20mA inputs. The ones I currently use have Johnson A350P's and I vary the signal with a potentiometer. When I turned in the purchase requisitions I was told that they had used resistors and resistors only to convert the 0-10 to 4-20 in the past and I didn't need to buy a converter (A350P) to do it.

I thought that changing 0-10 to 4-20 required an op-amp in addition to resistors, but I was told it had been done with resistors only on a 0-10V input with a potentiometer to vary it. I tested the old test stand that this method was used on but I couldn't get it to work. When I explained that to my boss he told me the old one worked. If it did I have no idea how so I would appreciate any help or expertise anyone that is more electronically minded could offer.

Thanks

Robert

Nope! You could get 0-1 mA or 0-20 mA but getting 4 of anything from 0 volts isn't happening. Naturally, loading a 0 - 10 volt supply with 10,000 or 500 ohms, would get you 0-1mA and 0-20mA respectively. That isn't how it's done in the real world because they need to look into varying load resistances and deal with the resistance of the connecting wires. You need an active output to be able to drive the output voltage higher as load resistance increases, to get the same current out for a given voltage in.

Now I have personally fed a device with a 4-20 mA OP into a SCADA analog input looking for 0-1mA with a 1000 ohm burden resistance. This was done by a combination of parallel resistance and computer programming.

At least, voltage to current is easy to do WITH an op-amp. If you give up on the passives and want to try a circuit, or explain your source and load specifics, we can solve it.
Joe

The signals go to a DDC (DX-9100) so that we can simulate transducers or transmitter signals. The 0-10V has a 300 Ohm impedance and the 4-20 has a 100 Ohm impedance.

We usually have potentiometers to vary the signal so that we can test staging of fans, compressors, etc. If I could build an op amp circuit cheap then they will approve the project.

I will try an figure out how to post a couple of PDF's of what we currently use and what I reverse engineered from the old test panel and put them up.

Thanks.

It's actually very easy. I was stuck baby sitting a substation today so I drew up a circuit. Mine included some output resistance, transorb, and fuse protection. This sounds like stuff you don't need. My designs often translate signals to drive a SCADA input 1000's of feet away, so I have to protect them.

OK, the way I see it, you need.
1.) 1, 24VCT transformer,give or take.
2.) 1 bridge rectifier.
3.) 2, quad op amps.
4.) 1, pos. regulator
5.) 2, neg. regulators
6.) 1, 625 ohm precision resistor.
7.) 8 pots, multi-turn if nec.
8.) A hand full of filter & bypass caps
9.) Something to build it in or on.
10.) Line cord, input fuse, switches, VM, mAM, if you aren't using a DMM and need to read values.

OK, all this stuff just counts if you wish to have 4, low impedance 0-10 volt sources and 4, precise 0 to 10 volt to 4 to 20mA converters. One quad op amp would be set up as op amp followers and the other as 4 V-A.
But if you don't need all this stuff and just want to drive inputs, go to the next post. I'll be doing a couple of drawings and submitting them.
Joe

[This message has been edited by JoeTestingEngr (edited 08-01-2006).]

But if you just want to ramp up your inputs and watch them with meters and/or scopes...

Stable 10 volt supply, 8, 2K pots, 4,400 ohm resistors.

4-20mA with 100 ohm load:
Place 400 ohm resistor in series with 2000 ohm pot in series with the load. Tie the wiper of the pot to one end so clockwise rotation reduces resistance. Place the series network across the 10 Volt supply.

0-10V with 300 ohm load:
Place ends of pot directly across 10 Volt Supply. Run wiper to top of load, bottom of load to negative of pot and PS.

Maximum PS load = (4 x 20 mA) + (4 X (10/2000)) + (4 x 10/300) = 80 + 20 + 133 = 233 mA. So a 10V, 250mA supply would be enough if all pots were full clockwise.

Joe


[This message has been edited by JoeTestingEngr (edited 08-02-2006).]

Thanks to Joe for the help. I will try to send in pictures of my test stand if it ever gets approved and built.

Don't build it.....

Phoenix Contact makes a small DIN rail mounted module that will convert just about any millvolt and/or amp signal to any other millivolt and/or milliamp signal.

Check out this link: http://eshop.phoenixcontact.com/phoenix/logon.do?user=anonym&general=enen

And search for part 2810913 MCR-C-UI-UI-DCI

Using DIP switches it can be set up to change just about any signal into any other signal. It's very nice. No idea of cost but I was about to buy one to solve a problem that I have. I think they run a couple hundred bucks....

I think they also make a unit that can be programmed for unique input and output performance.

RSlater,
RSmike

And search for part 2810913 MCR-C-UI-UI-DCI

DigiKey only showed the MCR-CLP for > $350.

Mouser shows the 2810913 MCR-C-UI-UI-DCI.

They want $400.26 for one and $358.85 each for 5 or more. So he could get the 8 he would need shipped to him for just under $2900. I'm guessing this would bust his budget. Of course he would have to get DIN rails and a proper PS. He wouldn't need the isolation and all the configuration options. He would still have to buy pots to drive the Mini MCR so that's a wash. He would still need a 24 volt supply that would possibly cost a buck or two less than the ckt I sent him, so that's almost a wash. The hardest thing to find was 625 ohm resistors. He would have to buy 620s and 4.99s or 5.10s but he would still have < 20 bucks in precision resistors.

So my reality is that I could have what Robert needs built on a PCB in my lab, in 1 day, with most parts on hand. Your way would cost me several days of frustration, dealing with Purchasing, WBEs, DBEs, 30% markups, and most likely getting something other than what I ordered. Of course, if we contracted out to build it....5 years and 10 million bucks!
Joe

Oh, and by the time it shows up, you've probably forgotten what you wanted it for.

Price is the issue in my case. I have a budget of $650 max. That includes sheetmetal, casters, wiring, etc. Luckily I am well known for scrounging up parts. That price does not include the DDC or any parts I already had on hand.

To put it this way, they were going to kill the project over eight A350P's at $70 a pop. However, I don't skimp on pots or precision resistors and the wire I have for it is leftovers from a job and it is high quality, high priced hook up wire.

Getting it going is priority one and if I have technical difficulties I have the leverage of "It's already 90% complete so cough up some dough to finish it."

It usually works. A mix of tenacity and persuasion can get alot done.

Sorry I thought this was more of an industrial application.

You have a lab???!!! I am so jealous. They took away my soldering station.....

Phoenix Contact also makes a unit that only does VDC to MA for about half the cost of 'one for all' module. About $200.

Back in the day I would have built it too but today I only play with big wires and voltages so I cannot justify my time on building something I can buy off the shelf for a few hundred dollars.

RSsorry,
RSmike

RSmike, I D/L'd several data sheets from the site that you mentioned. I wouldn't hesitate to use them if I had a control cabinet with DIN rail mounted components and a suitable PS. I appreciate that you posted them.

I emailed a couple of active and passive solutions to Robert and to Scott for posting I would be happy to fwd to you if you Email me, so we can all be on the same page.

Robert's needs seem to be more of factory bench testing with up to 8 changing inputs. I envision a panel with 8, linear pots. I would probably also use 8, NO PBs feeding a common wire to a DPM, so you can read the voltage on any wiper duing setting. The pots would all be fed from 0 and +10V. The V-Vs would feed Op Amp followers and the V-mAs would require the precision 625 ohm resistors going to a precision -2.5 volt reference.

The lab/basement is nice to have. That's where all the parts live. I transfer, etch, tin plate, drill, screen, spray, bake, stuff, and test the PCBs there. But the basement only has a dehumidifier so I layout and output to toner transfer paper from the office with the A/C.

How many times have you found a supplier who makes "almost" what you need? By the time you go through the hoops to make it "exactly" what you need, sometimes it's easier to build it.
Joe

Every time!

By the way, there are four ways to get 625 ohms using standard EIA-96 (1% or better) values in series:

475+150, 523+102, 324+301, and 316+309.

Also, 16.9K in parallel with 649 gets you pretty close.

(I got tired of doing this stuff by hand a long time ago, and wrote some utility programs to find the best series, parallel, and ratio values.)

If you are handy with a soldering iron, you could build your own V/I converter using a chip like the XTR110 from Burr-Brown: http://kolter.de/xtr110.pdf

Because you need to source 4 mA of current with zero voltage input, you will NEVER get there with a purely passive network of resistors.

[This message has been edited by NJwirenut (edited 08-20-2006).]