HX Autofill: Erratic behavior - flickering - Page 2

Equipment doesn't work? Troubleshooting? If you're handy, members can help.
User avatar
Randy G.

Postby Randy G. » Mar 14, 2019, 1:42 am

Is the probe possibly too long and/or is the heating element warped and touching the probe intermittantly...?
Espresso! My Espresso!
http://www.EspressoMyEspresso.com

Yacomonti

Postby Yacomonti » Mar 14, 2019, 2:29 am

Hi Jake!
Thanks, i liked the process to isolated de problem.

Here is my feedback, some of the answers are yes or no, depending the scenario (wich is weird). I just re do all of them based in this test procedure.

Jake_G wrote:•Does water touching probe shut off pump reliably?
-Yes: Probe is working correctly and boiler is grounded. Go to next question.
-No: Probe is not reading water, or boiler is not grounded. Fix probe or boiler

In a closed circuit:No, system flickering.
But if i baypass the selenoid: Yes.

Jake_G wrote:•Does removing the wire from the probe result in proper operation of the pump and solenoid?
-Yes: The controller, pump and solenoid are all working properly, go to next question.
-No: You got problems ;)

Yes always.

Jake_G wrote:•Does lifting the probe out of the water result in proper filling of the boiler?
-Yes: The probe is isolated from the boiler and apparently your machine is working. Yay.
-No: The probe has leakage from the wire (which works when disconnected if you passed the above tests) through to the boiler ground when the probe is not touching the water. Insulating sleeve is not doing its job...

In a "closed/normal operation" circuit: No, starts flickering when i lift the probe.
In a "open scenario/Bypassing the selenoid": Yes. I refill the boiler trought a pipe directly from the motor output and the autofill stops reading from the probe. No flickering by lifting the probe, refill system working with a raised level probe.


By the tests results, it seems that the cause of the malfunction could be the selenoid failing at opening or a leakage somewhere in the probe or cable.
I can confirm based in the procedure:
Selenoid and refill boiler working erratically in a "closed/normal operation" sensing boilerwater with the probe raised.
Selenoid and refill boiler works OK in a "closed/normal operation" disconecting the cable out and touch the boiler.
Selenoid and refill boiler works OK in an "open/selenoid bypass" sensing boilerwater with the probe.


Could be a poor grounded selenoid or something like that when the ground is read from probe or is impossible?
The ground connection is made by 2 main cable, one go from the Machine Case to control box and motor, and the other from the Case to selenoids and heating element.

Yacomonti

Postby Yacomonti » Mar 14, 2019, 2:37 am

Randy G. wrote:Is the probe possibly too long and/or is the heating element warped and touching the probe intermittantly...?


Hi Randi!
Just test it, i introduce a tick pencil which is the approximate length of the probe, and i think didnt reach any palpable obstruction.
I will take the element away to confirm.

Thanks!

Yacomonti

Postby Yacomonti » Mar 15, 2019, 1:49 am

Hello guys!
I made simple video to show the tests procedure and the behavior described in the lastest post in response of Jake base procedure.


Yacomonti

Postby Yacomonti » Mar 30, 2019, 4:22 am

Thanks for all the help guys, you give me a very enlightening theoric learning for the analysis.
It was a problem with the control box.

Since the autofill worked with and empty boiler and the rapidly on and off behavior of the autofill happened with the probe, and touching the boiler also activated the autofill ok, there was a 80% probability that were a ground reference problem with the box.

First I rewired all the ground cable line, but still the same issue.
I have a broken multimeter :s, so I change the components that could manage voltage in the lines of ground reference and sensor.
I replace 2 transistor, 5 resistances and the Integrated circuit.

The best of this thinks isn't repair things, the real satisfaction is the knowledge earned in the analysis process.

Cheers!