Help -What switch is this

[KeepItSimpleStupid]

I believe your correct where the ball goes. So, you definately have some sort of rotary encoder. The 0 ohm resistance confirms that the contacts are not resistive. If you don;t have dieectric grease, a thin coating of petroleum jelly to lubricate the contacts (the entire brown thing) would be good.

Any replacement is like to be optical. This doesn't men you can't get an optical encoder to work.

So how many positions again?
So, you have 20 divits and each time the ball goes between them, you get a detent?
Continuously rotatng, correct?

My hunch says that with this encoder you can't tell direction?

Again, I won't say this https://www.sparkfun.com/products/10982 would work, but it gives you an idea of what "might" be able to work.

It just doesn't look like you can figure out it's direction. So, one of the outputs from say the encoder above "MIGHT" work?

Do you have a pic of the PCB front and back of the board tat this came out of?

So, the two terminals annotated with 'white lines" are connected together (The two end terminals)?
If so, you should see the same change in voltage between the end terminals and the middle terminal as you rotate. I still can't figure out how it might know the rotating direction or does it need to know?

 

[stuhagen]

rotations controls 5 fan speed positions. Eve though it might be called a 20 detent switch
When rotating the switch it detents in those divits as you can feel in your fingers as it is rotating.
Yes the knob can spin endlessly or 360. But once past 5 in either direction it doesn't matter and oddly one return the other direction and the fan speed changes.
It has to know direction in either CW or CCW as one tells the switch to up speed or down speed.

I will attach the overall schematic if you want....but I had a thread about this over a year ago https://www.electro-tech-online.com/threads/heater-fan-switch.141100/

I believe your correct where the ball goes. So, you definately have some sort of rotary encoder. The 0 ohm resistance confirms that the contacts are not resistive. If you don;t have dieectric grease, a thin coating of petroleum jelly to lubricate the contacts (the entire brown thing) would be good.

Any replacement is like to be optical. This doesn't men you can't get an optical encoder to work.

So how many positions again?
So, you have 20 divits and each time the ball goes between them, you get a detent?
Continuously rotatng, correct?

My hunch says that with this encoder you can't tell direction?

Again, I won't say this https://www.sparkfun.com/products/10982 would work, but it gives you an idea of what "might" be abe to work.

It just doesn't look like you can figure out it's direction. So, one of the outputs from say the encoder above "MIGHT" work?

Do you have a pic of the PCB front and back of the board tat this came out of?

 

[Les Jones]

Hi stuhagen,
In post #14 you say the resistance reading changes in steps but then slowly rises when rotation is stopped. Was this test done with the switch/potentiometer connected to the controller or with it not connected to anything other than the multimeter ? I am thinking that maybe rotating it clockwise just gives pulses from one of the side contacts and rotating it anticlockwise just gives pulses from the other contact. If it was connected to the controller there could be a capacitor connected to it which may explain this behaviour. Can you post the schematic of the controller board or if you do not have the full schematic just trace out the circuitry around the switch/potentiometer input.
In the lower picture in post #1 I suspect that the radial lines between the two circular tracks originally joined the two circular tracks but have been worn through with friction with the fixed contacts or the ball.
It is also strange that there is nothing to control the path taken by the ball or retain it in a fixed position. Can you post a photo of the other face of the top item in post #1 ?

Les.

 

[KeepItSimpleStupid]

You said two switches worked. That's good. The encoders are only rated for about 10 mA @ 5V.

Look here: https://www.onlinecomponents.com/parametricsearch/electromechanical_encoders_mechanical-encoders.html?manf=alps electric&fixed=true

Probably something will work. Basically, it looks like you have a mechanical encoder. maybe with a little work, you can use an optical encoder and be done with it?

If yu know nothing about the circuit, we can probably use a couple of optomos relay's to replace the mechanical encoder. But, that;s doing it the hard way. I assume there's a 5V power supply in the gizmo?

 

[alec_t]

Taking the latest measurements into account I agree this is an encoder rather than a pot.
Here's how I now think it might work:-

A and B are spring arms connected to pins 1 and 3 respectively but not to each other.
If the shaft is rotated CW, the ball will locate itself under hole C and at each detent will short arm A to the wiper (pin 2) briefly.
If the shaft is rotated CCW, the ball will locate itself under hole D and at each detent will short arm B to the wiper (pin 2) briefly.
As Les says, "In the lower picture in post #1 I suspect that the radial lines between the two circular tracks originally joined the two circular tracks but have been worn through with friction with the fixed contacts or the ball."
That wear would make the shorting action unreliable/non-existent. Any chance you could do some micro-surgery and reinstate the radial bits?

 

[zahwi]

Here's how I now think it might work:-

I agree with your explanation that the ball applies momentary contact from wiper to A when turning CW and momentary contact to B when turned CCW.
I think that the problem is that arms A,B that press on the ball got weaken in time and the ball dislodged. The solution can be bending the A,B arms downwards so they press the ball more and the ball stops being dislodged by the grease.
The picture at post #1 shows that there is rubber in holes C,D it could be used to apply pressure on the ball from the casing. It could be that the rubber got lost in the grease and that allowed the ball to drift.
BTW, if there were 2 balls then it would have been a typical rotary encoder, as it is, it's a special switch.

 

[alec_t]

If it were me I'd simply replace the 'encoder' with a SP Mom.On-OFF-Mom.ON spring-biased switch like this. Nudge it either way to give a brief contact closure. Not so aesthetically pleasing as a rotary switch maybe, but it should do the same job.

 

[KeepItSimpleStupid]

I actually have this https://www.sparkfun.com/products/10982 encoder somewhere. It looks as if it's a mechanical encoder too. You would need to use pins A, B and C. A, B and C is defined in the datasheet and 1,2,3,4,5 are defined in the dimensional drawing. If the pins line up, it should work.

Here's https://www.newark.com/bourns/pel12...T1500?MER=ACC_N_L5_SensorsAndTransducers_None another.

There were a fair number of ALPS stuff in the link I gave earlier.

Like I said earlier, but did not elaborate, a optical encoder and n OPTOMOS https://www.digikey.com/product-search/en?mpart=CPC2030N&vendor=212 relay and say something like this https://www.digikey.com/product-detail/en/C14D16P-B3/102-1921-ND/2039181 and two resistors, you would have something infinitely more reliable.

The relay and maybe a couple of SMT resistors could be mounted "dead bug" style anywhere. The transistors would drive the LED in the OPTOMOS relay, so the resistors would be sized for that. The ON resistance is about 30 ohms.

This https://www.cui.com/product/resource/c14.pdf doesn't have the rating of Ic, however.

 

[stuhagen]

2 years ago when this first started to rears its head, my buddy who worked for Toyota (and speaks Japanese) told me that this switch basically sends a momentary 12v out to pins 1 and 3. This power signal to the main AC controller module tells it to change speeds. So like stated above, I went and mocked up two push momentary switches replicating the switch. By pushing one switch the fan operation for down worked and the other switch worked for up. So this switch must be a 2 pole momentary switch. Why I get odd resistance measurements I do not know. I do know Pin 2 is a constant 12v that momentarily (by rotation) sends it to the AC control module to change fan speeds. It must somehow send the 12v pulse and releases quickly. Which might be all about the internal springs/paths/ball...etc. Also agree that by taking a close look via magnify glass it is pretty pitted and worn. Even ALPS gives a lifespan of "xxxxxxx" rotations before it might go bad.

 

[KeepItSimpleStupid]

That makes more sense. There are chips that can create up and down from quadrature. e.g. **broken link removed** and https://www.elmelectronics.com/ebench.html#RotaryDecoders

So, the lack of a 5V supply is an issue. You still can interface using the OPTOMOS relays.

 

[stuhagen]

I made an error showing 5v originally....That was for the temp circuitry. I am limited to real-estate inside this box. I need something that can fit in the same footprint of the existing switch.

 

[KeepItSimpleStupid]

Your pic here, https://www.electro-tech-online.com/attachments/rotary-switch-jpg.86076/ suggests that there might be some room for a few dead bugs (IC's glued upside down).

It's great that +5 is available and supposedly +12 too. So,, if you can find an optical encoder that fits the same footprint (thread, attachment) your 80% there. Th interface would be "dead bug" style on the component side of the PCB.

The pics here **broken link removed** are incredible.