accuracy of testo temp clamps

[ats890]

Anyone have issues with the testo 550 temp clamps not being accurate in pac
kage units seems like they not well insulated from airflow

[Milk man]

I've had doubts also.

[mark beiser]

The sensors themselves are very accurate.

The problem is that to get precise readings, the sensor needs to be making good contact with the pipe along the full length of the sensor barrel, and if there is more than a few degrees temperature difference between the pipe and the ambient air, you need to insulate it.

[jim bergmann]

This is an interesting questions as there are many factors to consider.

First, the sensor is not more affected by the ambient air than the uninsulated pipe itself. If the pipe is exposed to ambient air it is absorbing heat also. When the fan starts the temperature of the pipe will change if there is only vapor in the line. Any uninsulated suction line should be insulated all the way back to the compressor. Any heat picked up in the suction line has to be rejected by the condenser reducing capacity, plus it reduces the cooling to the compressor.

Second, air is considered an insulator (if it is not moving) Shielding the clamp from the wind would be more important than insulating it from ambient temperatures.

Third, the exterior of the pipe is never the same as the temperature of the refrigerant flowing through it. I have tested several pipe clamp designs and found the best clamps are typically within 1 degree of the saturation temperature of the refrigerant. BTW, Testo and Digi-Cool use the most accurate sensors. I have tested them both. They are typically within .9 degrees of the saturation temperature.

Position of the probe affects the readings. Just like clock position of a TXV sensing bulb, a temperature sensor should be positioned the same way. I have a clamp probe tester that I built, and even on a 1/2" line there is a temperature gradient between the top and bottom of the line. If you look closely at the end of the sensor, that is the coldest are as the clamp is insulating the pipe from the wind.

Radiant heat could influence the reading. I have never understood why we do not make white Armaflex. I have found the clamp probe to be minimally influenced by solar gain.

The sensor mount can affect the accuracy. Testo has temperature compensation built into the Testo 570 for additional accuracy.

Last, because you are using the value to calculate superheat and sub-cooling you have to consider the total uncertainty of the instrument. The pressure reading will be converted to a saturation temperature. Any inaccuracies in the pressure measurement will also affect the calculation. If you leave refrigerant liquid in your hoses, clamp the probe to the hose and let it sit, the pressure in the hose will correspond to the saturation temperature of the refrigerant which should correspond to the temperature of the probes. I do this all the time for a quick check of the refrigerant and the refrigerant purity before hooking up gauges to a new system.

Testo uses atmospheric pressure (manual setting in the 550, and automatic in the 557 and 570). Digi-Cool can be adjusted in a vacuum. All manufactures of gauges today use relative sensors. If there is no way to compensate for the effects of atmospheric pressure the superheat and sub-cooling calculations will also be negatively impacted.

But to answer your question, I have found the effects to be very small as the pipe is a much better conductor of heat to the sensor than air to the sensor. The majority of the sensor is in contact with the pipe of the plastic insulator that it is housed in.

I attached a thermal image of a Testo clamp probe strapped to a pipe. You can see that the sensor is very close the the pipe surface temperature. If we were to blow air on the assembly, the temperature of the entire assembly would be affected.

[ats890]

Good info Jim I feel more confident in trusting its readings just have to make sure clamp is on pipe correctly and let 550 do the rest

[jim bergmann]

I shot a couple of videos that you might find interesting on clamp probe accuracy.

http://youtu.be/k9LXhw8WVU0

http://youtu.be/1Llh4PaaGnI

http://youtu.be/mtsQ2X07f0A

[itsiceman]

Nice video Jim. You up for placing the temp. sensor directly to the pipe wrap some tin foil around the sensor and pipe then wrap insulation over the whole thing?
Copper conducts very well IMO and I have seen numbers matching up under saturated conditions with different meters and sensors (523 and 1250)
I believe you might see the same.

[jim bergmann]

Nice video Jim. You up for placing the temp. sensor directly to the pipe wrap some tin foil around the sensor and pipe then wrap insulation over the whole thing?
Copper conducts very well IMO and I have seen numbers matching up under saturated conditions with different meters and sensors (523 and 1250)
I believe you might see the same.

Sure, I can shoot that tomorrow. I have some heat sink compound also I can add to the mix. Considering however that it takes about 2 seconds to install the clamp, I could live with 0.7 to 1.5 degrees. I will try your suggestion though, it will be interesting.

[itsiceman]

You would need a little more time rigging up the clamp to work in a package unit like the OP noted ;)

[jim bergmann]

Ice,
Here you go.

http://youtu.be/B9cyDrQ3xTU

[itsiceman]

looks good bro :putergreet:
the tin foil was for under the ins. though lol
it carries the temp of the pipe up and around the hole barrel and will give you a quicker response to temp changes :.02:

[chuckcrj]

Very good illustration Jim!

It is interesting that a temp clamp can be over 1° off in a 78° room. I wonder how far off it would be on a 90° roof in direct sunlight while clamped on a 45° suction line?

[jim bergmann]

The way that the testo clamps are made with the foot that is separate from the clamp body, I am guessing not a lot different. If stuck in direct sunlight though a piece of tin foil over the clamp might really help with additional accuracy.