Hi all, First post so go easy!

I have looked through countless posts on this before asking but none seem to be straightforward or else i need them to be explained more easily

I need to develop a 4-20mA supply to fed an inverter for a project.
The inverter has a 24v output on it i can use to power the circuit or i can strip down an old mobile charger to give me 5v or i can make a PP3 9v batttery powered unit

I have been given an XTR110kp and some knowhow of circuits and components. What i dont know is how to do it.

The datasheet says 10v or 5v Vin but also a Vcc of upto 40v.

Can i just power up the chip with 5v to Vin2 and use variable resistors accordingto the datasheet values to generate the 4-20 or do i need the 5v to be variable?

If the inputs need to be a variable 0-10v (Vin1) or 0-5v (Vin2) how do i do it? with what i've got to hand.
I can get pots and resistors and other small components readily enough but dont want to have to buy a variable power supply or expensive components.
I do have couple of Fluke multimeters etc so can use them to measure mA and voltages etc

 

The datasheet says 10v or 5v Vin but also a Vcc of upto 40v.

Can i just power up the chip with 5v ...

No, the datasheet clearly says the device needs min 13.5v. The 5v (or 10v) input is what will be converted to a mA signal. You can use a pot and a resistor voltage divider to vary the Vin voltage.

It sounds like you plan to control an inverter using a 4-20mA current loop? Unless you have a special inverter made for that purpose, I don't think that'll be easy. What exactly are you trying to do? There are some clever folks here that can help you, if you're willing to abandon preconceived ideas. Hard sometimes, but that's how we learn.

 

Yeah i have just seen the part re the Vcc range but needed some clarification.

Why do i need to vary the input 0-5v or 10v if the output has pots to adjust the span and gain. Does a fixed input voltage not work?

My Inverter does have the facility to be controlled using a 4-20mA input. It can also be controlled using a standard Pot ,a series of inputs from a PLC program to give set speeds, a 0-5v and a 0-10v input but i want to be able to do all of them. Its about being able to control an inverter using a variety of inputs and unless i try to do all of them then i'm not learning anything.
The easy way out is to buy a 1K 2W pot, connect it in and away you go. But i still wouldn't be any wiser or able to control it if it was in an instrumentation loop.

I'm getting there slowly but still need help.

I initially thought of just connecting a 1200ohm resistor in series with a variable of at least 4800ohm to give the 24v a 4 - 20mA range using V=IxR. and using a fluke in series to measure the current. IS it that simple ?

 

I initially thought of just connecting a 1200ohm resistor in series with a variable of at least 4800ohm to give the 24v a 4 - 20mA range using V=IxR. and using a fluke in series to measure the current. IS it that simple ?

yes in theory. - at 20ma that's .48W so I would use 1W resistor & pot, if not more to be safe....but, I would think that the 24V output from the drive is meant for digital logic (relay contacts/pushbuttons) making use of it's own high impedance inputs, so it might not put out enough to dissipate 1/2W across your pot & resistor. it would probably be better to use the 5V p/s you mentioned since you would be able to use lower wattage resistor/pot also.

 

Why do i need to vary the input 0-5v or 10v if the output has pots to adjust the span and gain. Does a fixed input voltage not work?

Maybe I'm missing something but I thought the whole purpose of that fancy IC was to convert an input voltage (either 0-5 or 0-10v range) to a 4-20mA current output. If you're not doing that with the IC, why use it?

That conversion is very handy by the way. I built a driver for a LCD panel meter that uses 4-20mA. My driver uses op-amps to convert a small voltage (0-1V, if I remember right) to that 4-20mA signal. Having this fancy chip of yours would have made for a very much simpler circuit.

 

Sorry bout the confusion, I'm not saying i'm not going to use the chip. I'd rather use it i'm just limited in what i have, and a 0-5 or 0-10 v source is one of those things i'm lacking in. but open to ideas

 

Let's back up and talk about what you really want to do. As I understand it, you have an inverter you want to control (output voltage? on/off?) and that inverter has a 4-20mA current loop control capability. Right?

Now, you'll want to control your inverter by turning a pot. So somehow you need a circuit that will deliver 4-20mA through the inverter's control circuit. Right?

The xtr110 is an excellent choice for this, since it will linearly convert a voltage (with almost no input current) into the appropriate 4-20mA output current. All you need to input into this chip is a voltage reference, and a resistor voltage divider with a pot to dial in the exact voltage - and thus current output - you want. Or you can take a voltage output from a controller directly into this chip.

There are other approaches that aren't bad, using op-amps to again convert a voltage input into the required current signal. The xtr110 is a specialized version of this, but it can be accomplished with basic components.

I believe the least favorable approach would be to directly drive a current through the loop using some combination of resistors and a pot. You MIGHT find a combination that works fine for you, but you'll need to experiment quite a bit to get the values right. And I wouldn't count on having smooth control over the entire range. Worse, you might discover the hard way that you've overloaded the inverter input.

One final point. If I recall correctly, my 4-20mA panel meter required a voltage of ~10v across it to achieve a 20mA current. If your inverter is similar, you'll need a supply voltage at least that high, or you'll never achieve the full range. Using the xtr110, the CONTROL voltage can be 0-5v, but the xtr110 or other approach will require a higher voltage SUPPLY, in order to drive the needed current at the output.

 

I believe the least favorable approach would be to directly drive a current through the loop using some combination of resistors and a pot. You MIGHT find a combination that works fine for you, but you'll need to experiment quite a bit to get the values right. And I wouldn't count on having smooth control over the entire range. Worse, you might discover the hard way that you've overloaded the inverter input.

I don't see why this wouldn't work. I should test this in real life to learn where you are coming from, but I don't have a spare inverter around. to me in theory it seems pretty straight forward. Would you mind explaining so I can understand? my thinking is: If he used a 5V external power supply, a 250Ω series resistor & a 1kΩ pot, that should be 4-20mA right there. Why the chip? Those multifunction inputs probably have a switch for 0-10V & 4-20mA which puts a XXXΩ resistor across the terminals, are you referring to error caused by loading on that?
Thanks wayneh

 

If he used a 5V external power supply, a 250Ω series resistor & a 1kΩ pot, that should be 4-20mA right there.

That's ignoring the internal resistance of the inverter, which is both unknown and probably not constant. The chip corrects for all that by giving a linear conversion from input voltage to output current.