Must unplug and replug power to make Cooker Hood / Kitchen-Fan work

[carlsweden]

Problem:

Nor the fan nor the built-in spotlights will work. Then I unplug the power plug, wait for 5 min, and re-attach it to the power socket. Then both fan and spotlights work when switching them on using the normal buttons on the cooker hood. But after switching them off using the same buttons, and then trying to switch them on again, they will usually not work until I again detach and re-attach powerplug as described above.
When looking and smelling the circuit board I do not see anything out of the ordinary.

Question:

Do you have any idea what the problem may be? If so, is it something I can fix myself? If so, how?
Would it be a high chance of solving the problem by ordering a new circuit board from the manufacturer? (Circuit board (excluding transformers) costs 150 euros. A complete new cooker hood costs 400 euros)

Manufacturer:

   Franke (The actual brand is Cylinda, but on the inside it says Franke Futurum)

Model:

   Cylinda Sunda RF, but I believe the insides is the same as Franke’s model: F600 60 RF Franke.

Year: 2007

I attach PDF with an electrical diagram and photo of circuit board.

Any help is much appreciated.
Thank you!
Carl

[abraxa]

It's ambiguous whether they only stop working if you had *both* running or whether they *both* stop running even if only the fan or light was used.

So to clarify: if you plug in the device and switch on the light, then off again - do the lights still work or do they stop working? What about the fan?
Also, if you plug in the device and switch on the fan, then off again - does the fan still work or does it stop working? What about the lights?

Does running the fan for a short time make any difference (i.e. only 2 seconds instead of a minute)?
Did the device always behave this way or is this behavior a recent development?

That said, a better picture of the PCB (and its back side) would be good. My personal guess would be that the microcontroller goes haywire when one of the relays turns off.

[eKretz]

Also, if you plug it in and first leave it alone for 10-15 minutes,  do the fan or light still turn on after it has been sitting unused while plugged in? We need more diagnostic info.

[IanB]

Something similar happened to a cooker hood/fan belonging to a relative of mine. It would stop working unless you disconnected the power and reconnected the power to reset it. Apparently the problem was something to do with the control board getting latched into a bad state. I believe replacing the control board fixed it. If the whole control board is too expensive it might be possible to replace a single part on the board, but if the part that is causing the problem is a microcontroller I don't know if it could be replaced while preserving its original program.

[borghese]

Have you checked the power supply?

[Seekonk]

I'd solder two wires to the capacitor and bring them out so the voltage can be monitored. If the capacitor weakens, the micro will continuously reset.

[drussell]

Wow, talk about something with an unnecessary amount of overly complex electronics to do something traditionally handled by a couple of mechanical switches. 

[james_s]

Agreed there, this is a classic case of something being designed to be more complex than it has to be, likely in the name of having trendy high tech electronic controls.

The first thing I would suspect is capacitors in the power supply, as was suggested already there's probably a microcontroller in there and the program is crashing. This sort of thing can happen when the capacitors dry up and the power supply voltage goes out of spec or develops a lot of ripple.

[carlsweden]

Thank you so much to all of you who replied to my question!
I bought this for my mums 60 year birthday several years ago, and the price of a new one is 2 weeks salary for me, so I soo very much appreciate you guys trying to help me. Thank you!

I have made more detailed testing based on your questions. Below I write the results.

How it used to work before problem started:
The cooker hood has 5 touch-buttons (photo 1) that connect directly to the circuit board via a ribbon conector.
1 button for the spotlights. 4 buttons for the 4 different fan-speeds.
When pushing a button the first time, it makes something (I assume it’s the relays) on the circuit board make a “clicking”-sound, and then the function that belongs to that button is initiated (for example lights go on if it’s the light button).
The second time the button is touched; it again makes the clicking sound, and stops the function that corresponds to that button (for example switch off the lights).

How it works now after the problem appeared 2 weeks ago:
1.   If I plug in power supply and wait without pushing any buttons for 1 minute or more, nothing works when I finally push the buttons.
2.   If I plug in power and within seconds push one of the buttons (light or fan), it will work.
a.   If I then switch off the function, then I will not be able to switch on any function without first unplugging and replugging powercord. (Even if I only ran the fan or had the light on for 2 seconds)
3.   While having light on, I can switch the fan on and off and switch fan speeds as many times I want without malfunction.
4.   While having one fan speed on, I can switch light on and off as many times I want without malfunction.
a.   However, if light is switched off, and I have one fan speed switched on and working, then if I push the button for another fan speed, then everything will stop working (each speed-change involes pushing the fan-button that corresponds to that speed and have the corresponding relay “clicking” sound.
5.   Sometimes, when I push the button for fan speed number 2 (after having disconnected power just before to make it work), the light will come on at the same time without me pushing the light button. This happens maybe one time out of 4 tries. This never happened before the other problems started.
6.   Above results do not change when I try different power outlets, and I have verified that the Halogen spotlights do not exceed the max Effect (20W). Also, the fan has been regularly cleaned inside and appears to run smoothly.

Questions:

1.   Is there anything I can try to fix the problem?
2.   Is there a high probability of resolving the problem by buying new circuit board from manufacturer? (cost 150 euros).

[james_s]

The very first thing I would try doing is replace that electrolytic capacitor. You could also poke around the microcontroller (Atmel chip) and check that the right voltage is present and the clock signal is there.

It's very unusual that they have helpfully brought out a bunch of signals to test points at the edge of the board, looks like the power rails and ISP programming signals as well as some IO pins. 

[carlsweden]

Thanks James!
I have basic solding knowledge, so can change those if you tell me which one it is?
Is it the 6 black square capacitors called omron g6rn-1 12vdc 10a in the picture?
Is it one in particular? Or should I change all 6?

[james_s]

No, those Omron things are the relays that control the motor and lights.

The capacitor is the cylindrical object in the lower-right of the picture. I can't read what it says but normally it will be rated in microfarads (uF) and a max voltage. Probably something in the area of 470-1,000uF at 16-25V.


Do you have a multimeter? See those pads along the lower-left that are marked Vcc and Gnd? Connect power up to the board and measure the DC voltage between those, I would expect something like 3.3V or 5V, and it should be stable. If it's jumping around, or way outside that range that's a good clue something is wrong.

[oventech]

See the PCB board is not flow soldered. Just surface mount devices and the thru hole parts added to the board. Wonder if any of the via's are oxidized. Have seen failures due to no or intermittant connections between the top and bottom sides of the board. Check the power supply voltages.

[johnkenyon]

What's the markings on the yellow device to the north of the transformer near the markings "FAS NOLL"?
Is there any evidence on the top of the board that it might be getting warm?

I'm thinking that it could be either a polyfuse or a PTC thermistor intended to protect the transformer.

Might be worth measuring the AC voltage across it when turned on/cold/working, and comparing the voltage when its turned on/warm/failed. (Multimeter set to 1000V AC range in both cases).
If the voltage across the device goes up, then either the device is being triggered by a power supply fault, or the device itself is faulty.

/John

[james_s]

That looks like a MOV to me, although I can't see how it is connected since the top side trace(s) are blocked by other stuff.

[eKretz]

I would try that capacitor too. If everything's working with the light switched on perhaps the bulb is acting somewhat like an inductor and dampening the fluctuating voltage from a failing cap. That could be causing something to go a little haywire (M/C?) after a short duration when the light is off.

[carlsweden]

Again, thank you for your input.
I just received the capacitor I ordered and am exited to solder it onto the circuit board, BUT, there is one question first:
There is a sign on the new one that is not present on the old one. Could someone have a look at the photo and say if they are the same?
They are both 100uF and 50V, but the white strip by the negative lead is marked " - - - " on the old one, but marked " - < - " on the new one I bought.
Are they interchangeable, or are they different, given that one has the "<" sing and one does not?
Photo attached.

Thanks!

[james_s]

There are many different styles of marking the negative side. The positive lead is also usually longer to indicate which one that is. Don't worry about small differences like that, they're the same.