Electronics > Repair

Zodiac Tri Pool Chlorinator Repeated Zener Diode Failure

<< < (4/6) > >>

soldar:
Things which would help:

- Bigger zenner (already done)

- Bigger C41 - I would start by doubling or even tripling the capacity

- Small value resistor or inductor in series with L5/2 - Difficult to determine best starting value but you can use a scope to see how the current behaves and have it smooth out

An alternative to the last one would be to try a resistor or cap in parallel with L5/2

Frankster:

--- Quote from: soldar on December 27, 2018, 11:36:20 am ---Things which would help:

- Bigger zenner (already done)

- Bigger C41 - I would start by doubling or even tripling the capacity

- Small value resistor or inductor in series with L5/2 - Difficult to determine best starting value but you can use a scope to see how the current behaves and have it smooth out

An alternative to the last one would be to try a resistor or cap in parallel with L5/2

--- End quote ---

Thanks heaps for reply soldar!

Bigger zener was my very first and second thought also. I NEEDED a working chlorinator TODAY so I hurriedly replaced zener with a 5W version and my unit is again functional (but I still have a friends spare dud unit with same fault to mess with.)

Bigger C41 hey!? Interesting that reference design has a 1000uF specified where Zodiac only use 100uF... (my plan was to check C41 ESR against a new one and replace it anyway..)

Small value resistor or inductor in series with L5/2. Like the 10R R31 in reference design? Also interesting that reference design also has a 10k resistor in parallel with zener where Zodiac doesn't...

Circlotron:
Never had the problem but I did read the following:


--- Quote ---Under certain conditions (low reverse current), some of these things exhibit a negative resistance, and can actually form a relaxation oscillator in the correct circuit configuration. Take a look at the expanded portion of Figure 5 on Page 19 of On Semi’s “TVS/Zener Theory and Design Considerations Handbook
--- End quote ---

http://www.onsemi.com/pub_link/Collateral/HBD854-D.PDF

So the zener might be oscillating and taking *big* bites of current from C41. Maybe.

fzabkar:
I suspect that the normal operating voltage across the Zener is less than 24V. This is suggested by the low margin of the 25V capacitor. If the purpose of the Zener is to contain voltage spikes at power-on, then perhaps a 24V TVS diode might be a better fit.

Frankster:

--- Quote from: Circlotron on December 28, 2018, 11:34:58 am ---Never had the problem but I did read the following:


--- Quote ---Under certain conditions (low reverse current), some of these things exhibit a negative resistance, and can actually form a relaxation oscillator in the correct circuit configuration. Take a look at the expanded portion of Figure 5 on Page 19 of On Semi’s “TVS/Zener Theory and Design Considerations Handbook
--- End quote ---

http://www.onsemi.com/pub_link/Collateral/HBD854-D.PDF

So the zener might be oscillating and taking *big* bites of current from C41. Maybe.

--- End quote ---

My tiny brain hurts after reading that. I had never heard of that phenomenon. How would I go about proving  this could be occurring and then rectifying it?


--- Quote from: fzabkar on December 28, 2018, 07:35:03 pm ---I suspect that the normal operating voltage across the Zener is less than 24V. This is suggested by the low margin of the 25V capacitor. If the purpose of the Zener is to contain voltage spikes at power-on, then perhaps a 24V TVS diode might be a better fit.

--- End quote ---

I suspect you are right about normal operating voltage being less than 24V, as chip VCC is internally clamped to 16 V. I need to go get and fit another zener in my mates dud spare and take some readings...
A TVS diode was been suggested before...

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod