Test Equipment Issues

[neongreen]

I'm not sure where else to post this, so I'll post it here.

I am putting together a test kit that will go in a vehicle for on track testing and have run into some issues. A former co-worker put together a nearly identical kit some years ago, and it's still in use. I assume it works, but have not worked with personally. That equipment is at a track in Arizona, I am in Michigan so I can't inspect it in person.

With that out of the way, this equipment is using a VBOX3i as it's base, and i'm sending some analog signals to it. The problem I'm having is with some radiation temperature sensors (two of each: Keyence FT-H30 sensor head with FT50-AW amplifier). These sensors will read the temperature of surfaces they are pointed at, and output a 4-20 mA signal. Problem is, the VBOX analog inputs can't read a current signal, they need a 0-5V signal. So the signal is sent to a signal conditioner (Omega DMD4380) for conversion.

The sensor amplifiers have a display, and when the sensor heads are pointed at the same surface, they read within about a degree of each other. When I probe the current output from the sensor amplifier to the signal conditioner, the current matches expected output perfectly, based on the display on the sensor amp. Yet when I probe the voltage on the other side of the signal conditioner, the incoming current has not be correctly translated into a voltage signal. The signal conditioners have a number of switches on the side that allow you to configure the input and output ranges. I have checked the charts and switches at least a dozen times, I am 100% certain they are set correctly. I've also calibrated and checked the zero and span several times (built a calibration device with a 24v power supply and a handful of resistors to simulate the 4 and 20 mA signals). I called Omega and described my problem, they seemed to think I had done everything right, and they would be willing to inspect my signal conditioners, but that's a lot of down time... What's going on here? is there some sort of incompatibility between my sensor and signal conditioner? Did some component inside the conditioner get fried?


----------------------------------------
I could get so much done if I didn't have to go to work

 

[IRstuff]

Some data would be useful, i.e., run in the input from 4 to 20mA in steps of 1mA, and what do you get?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[neongreen]

So when I manually input currents from 4-20 mA (using a 24v source and a 10k potentiometer) the readings in the VBOX look perfect.

So the temperature sensor and the signal conditioner both seem to work just fine independently, but when I connect them, it doesn't work.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[LionelHutz]

Why the complexity of the converter? Use ohms law. A high precision 250ohm resistor can be used to convert the 4-20mA signal into a 1-5V signal.

 

[Mongrel]

I'm not familiar with your parts, so this might not be a lot of help. But... a thought - some automotive sensors will send a 0-5v pulsed digital signal. It seems from your post that these aren't like that, but perhaps it would be worth putting a scope on the setup to see if you're getting a nice even signal, or if spikes and jumps might be affecting the output?

 

[IRstuff]

I was referring to the actual sensor data output, i.e., does it follow some rule, arbitrary or otherwise, or is it random?

You could use a thermal bath to get different temperatures

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[neongreen]

Lionel, fair point. I didn't think of that.

Mongrel, if these are pulsed, we have a real problem. The users manual states it should have a linear 4-20 mA output.

IRstuff, gotcha. I only have sensor data output for a few points, but the mA output is dead-on for those points.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[neongreen]

I'm now wondering if I have it wired wrong somehow?
This is the sensor diagram

Sig + Pink
Sig - Blue
V+ Brown




And this is the signal conditioner diagram:

I have it wired as follows:

9 Blue
10 Brown
11 Pink


I've played around with it a bit, but with no luck.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[neongreen]

Ok now here's where it gets weird... and possibly a bit closer to solving this issue.

The sensor amps and signal conditioners sit next to each other (and 2 thermocouple amps) on a DIN rail. I noticed that when I unplugged sensor 1's signal conditioner output, sensor 2's reading in the VBOX jumped up about 7 degrees to 19 C(readout on the sensor amp remained unchanged at 22 C). When I plugged it back in, sensor 2's reading dropped back down to 12C. If I unplugged the signal output on sensor 2's signal conditioner, sensor 1's reading in VBOX jumped up about 3 degrees. If both are plugged in and I put my hand near them, one of the readings(I forget which) drops 2C.

The signal wires are all part of a pre-made signal cable that has maybe 6 or 8 pairs of shielded wires. The extra pairs were just trimmed off and left in the outer casing.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[LionelHutz]

Did you do any basic testings, like measure the power supplies while it was operating?

The conditioner probably doesn't have enough power available on it's loop power supply to power the temperature sensor. A loop power supply only has to supply a little over 20mA, maybe about 22mA. This means it would have about 0.33W available on the 15V loop power pin. A Quick google search shows the temperature sensor needs at least 1.6W to power it. Bottom line is never use a loop power supply on a converter for any purpose besides powering a 4-20mA sensor loop.

Get a decent quality 24V power supply for the sensors.

 

[neongreen]

Ok that makes sense. I'm new to all this so I appreciate your help. This morning I received some photos of the other setup that is out in AZ, looks like they were not powering it off the Signal Conditioner. I have a little re-wiring to do, hopefully that clears it up.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[neongreen]

Lionel wins! The power source was inadequate. I powered it off the main 12v power bus in the equipment setup, and tied the blue 0v line to the ground. This cleared up most of the issue.
the only remaining problem is that one of the sensor amps has poor accuracy on it's 4-20 mA output. It is offset a little from what the display on the amplifier shows. I suppose I can fix that in the VBOX software by tailoring the scale and offset to match the amp display, though I'm terribly annoyed at the poor accuracy.

----------------------------------------
I could get so much done if I didn't have to go to work