JLD612 temperature instability when using PT100 probes

[pkoken]

This is a real puzzle, I am beginning to think that my JLD612 is bad.

I have built a sous vide cooker out of a countertop roaster. This uses a 1500w heating element fed through a 75A SSR.

I started out with a Type K thermocouple which worked very well. Unfortunately the 1 degree resolution of the Type K seemed to lead to larger temperature swings than I wanted.... so I ordered a PT100 probe.

The first PT100 probe showed unstable temperature measurements. I checked this forum and tried bonding the shield on the PT100 cable to the JLD612 with no change. I assumed that the problem was that I was not using a waterproof probe, and I had somehow contaminated the probe with water.

I ordered a new style waterproof PT100 - http://www.lightobject.com/Premium-Stai ... -P681.aspx which I installed in place of the non-waterproof probe. Unfortunately, the problem still occurs! I will point out that the construction quality of this probe is superior in every way.

I have put together 3 very brief videos showing the symptoms, and the setup. http://flickr.com/gp/s0cal/jva287/

[richiem]

Please give us the range of variation, and whether you are using the Pt100 or Pt10.0 Inty setting. Also tell us which wire of the probe is connected to which terminal of the JLD612.

[pkoken]

Setting is PT 10.0

The range of variation seems to be +/- 5 degrees on a continual cycling basis (see videos in my original post).

The A wire is connected to terminal #8 and the B wires are on #9 & 10.


Untitled by S0Cal, on Flickr

[Tech_Marco]

What happpen if you operate it without the enclosure closed?

Marco

[pkoken]

Good question!

Video is here ->http://flickr.com/gp/s0cal/3z4431

As you will see... the problem remains the same.

[richiem]

My conclusion is that the JLD612 is bad. My hypothesis is that its constant current source output to the probe is noisy, but it is only a hypothesis. I would need to hook up a 130 ohm resistor in place of the probe, and look at the A wire connection with a high-sensitivity oscilloscope to confirm the hypothesis.

[pkoken]

My personal conclusion is that the JLD612 is bad as well. I put this thread here largely for others to be able to learn from.