DIY Auto Draw Request for Help

[DRoseum]

I built a very basic auto draw system using a PID controller (Inkbird ITC-106), a 3 wire PT100 Rtd probe and a stainless solenoid valve.

I am not using the PID output control, but just have it set on the simple on/off control mode that uses the hysteresis temperature and the Set Value (SV) to control the solid state relay.

Everything works well except my temperature perceived valve (PV) from the probe varies wildly by +/- 1.5 degrees when in boiling water. For example it jumps between values like 211.0 and 212.5 every half second to second. This becomes a problem when trying to draw off syrup as the unsteady monitored temp causes a lot of cycling of the solenoid and doesn't seem to be very accurate or consistent in drawing off actual syrup.

Has anyone experienced this? Any suggestions on how to correct it?

I have tried changing the "digital filtering" value on the PID input settings with zero change in behavior. That setting ranges from 1 - 20 and has no effect on smoothing out the process noise of the input temp.

Sampling period for the PID is factory set at 0.5 sec and cannot be change thru user settings.

Is this a common problem? Is it a probe issue? Sampling period issue? Filtering issue? Something else?

Thanks in advance

[Biz]

Just a guess, could be picking up 60Hz noise. Try grounding the probe to an earth ground. Or shut off any appliances in the area. I would think the digital filtering would handle that but worth a try.

Dave

[DRoseum]

Just a guess, could be picking up 60Hz noise. Try grounding the probe to an earth ground. Or shut off any appliances in the area. I would think the digital filtering would handle that but worth a try.

Dave

Do you use one of the 3 wires to ground or ground probe / shielding itself direct with new wire?

[Shaun]

Another thing you can look at is any wire connections. Soldering the connections is best if possible. Being as RTD's are only a resistor any other resistance in the "run" can cause this. You can also put the probe in a glass of ice water and check the resistance with a multi meter. It will read 100 Ohm's at 32F.

[darkmachine]

I use the Inkbird 106vh in my draw off, I can check my settings, i seem to remember some flakeyness, it's not an expensive pid. It did do the job though, saved me a ton of time testing and boiling. I made all my fine adjustments with a hydrometer before canning. I use the on/off and then switched to AT which seemed to work. I considered going to a different valve setup, but I am pretty sure I got it to the point where it wouldn't bounce and try to open, over and over. I did add a downstream ball valve so they draws are longer and it doesn't surge.

I had to go back an check, when i originally posted i though i was using on/off, but here is the video that set me straight.

https://photos.app.goo.gl/zaaUNd8vV9Z2eBnx5

[DRoseum]

Thank you all! Some additional troubleshooting has led me to believe its definitely a 60hz line noise issue. Will get a power line filter to clean that up.

Will also give using the actual PID functions a go after that vs the on/off setting. Much appreciated!

[TonyL]

When I built mine I used a step down transformer to get the 24 vac needed for operation. I installed the transformer in the enclosure for a nice, clean looking setup.... it didn't fly, as my readings would jump around also. I moved the transformer out of the enclosure and attached it to the support stand itself, and that did the trick.

[darkmachine]

It was easier for me to use the ssr to make a switched receptical on the side of my control box. I had a laptop power supply to power the valve. I wonder how the AC vs DC affects operation?

[western mainer]

http://mapletrader.com/community/showthread.php?35076-Modulating-Auto-Draw-off
Go here and read up good plans and it works well.
Brian

[BSHC]

It was easier for me to use the ssr to make a switched receptacle on the side of my control box. I had a laptop power supply to power the valve. I wonder how the AC vs DC affects operation?
I found my solenoid didnt draw enough current to require the SSR. I believe i initially wired it up but it would not work, unless i plugged in a light bulb to increase the current. So i eliminated the SSR and put a label on the receptacle to remind me of the current limit.
The problem i have been having is getting an inexpensive PT100 to live more than a few days.
I have ended up using a K thermocouple but it seems to vary a lot compared to my analogue thermometer. I have to make hourly adjustments to keep them in line.
I need to find a durable PT100 that wont empty the bank available in Canada

[DrTimPerkins]

I have ended up using a K thermocouple but it seems to vary a lot compared to my analogue thermometer. I have to make hourly adjustments to keep them in line.

Thermocouples, while very inexpensive, also tend to have quite high error limits. Standard Type K thermocouples are about +/- 4 deg F. SLE (special limits of error) Type K thermocouples are +/- 2 deg F. Both are really too high an error for determining the draw-off temperature reliably. A PT100 thermometer (with a PTR sensor) has error limits better than 0.01 deg F. HUGE difference.

Bottom line...best to avoid thermocouples for drawoff systems.