Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length.

[BarryM]

This problem relates to a table-top fairground carousel but it could just as well have model boat implications.


The carousel plays music as it rotates and this was provided by a tape unit until 30 years of use saw its demise. I decided to fit a MP3 player that was small enough to fit into the limited space. A suitable music track was downloaded and the unit was tested as working fine. The snag was that the music was triggered by a pressure switch which was impractical for the setup. To overcome this, I cut off the pressure switch and replaced it with a 6V relay. Power up MP3, switch on relay and all will be fine - or so I thought.


If I connect the switch wires directly, the MP3 plays. If I switch it via the relay, there is no sound at all. The relay checks out as working correctly.


Will somebody suggest where I am going wrong or if there is an alternative to a relay? Bear in mind that my electronics knowledge would be regarded as dim by a fencepost.  Help!!
Barry M

[HMS Invisible]

Such devices will have an electronic latch triggered by a tactile switch with momentary action.
I think your problem is caused by contact bounce & might be solved by a low value ceramic or polyester capacitor across the relay contacts.

[BarryM]

Thanks for that very quick reply. Would something about 50 microfarads be OK?
Barry M

[HMS Invisible]

That's well beyond what I was thinking. It's likely just a tactile switch pulling a processor pin to 0volt with an internal pull up of ~ ten to 100kohm with the minimum protection. So a tiny capacitance will be noticed and high value electrolytic caps should be avoided. If you had long wires you'd be safer with ~ 1kohm resistor. That on its own might even do the trick and should be tried first.
 I suggest one to one hundred nanoFarad and don't go beyond one microFarad.
The marking code for one nF is 102. One hundred is 104.

[BarryM]

Thankyou - I've plenty of resistors, I'll try that first.  The length of wires from MP3 board to relay is about 50mm with a two-pin plug at half-way.
Just a thought - snatching at straws? - could that plug be in any way involved?
Regards,
Barry M

[HMS Invisible]

To clarify, I meant a series resistance rather than a direct short by relay.
I suggest bridge the contacts with a resistor to try it before soldering.

Five cm length is not excessive and I don't think a connector would have any effect.

[BarryM]

Sorry but do you mean put the resistor across the relay contacts or put it in series with one of the switch wires?

[HMS Invisible]

Sorry but do you mean put the resistor across the relay contacts or put it in series with one of the switch wires?

In series with.
To visualise, scroll down to the first diagram here.
https://hackaday.com/2015/12/09/embed-with-elliot-debounce-your-noisy-buttons-part-i/
R1 might sort it on its own.
C1 might sort it on its own.
Don't worry if you soldered C1 across the relay. It doesn't much matter.
Rp1, 10k is internal to the microcontroller so ignore it completely.

[BarryM]

That's very helpful. I'll give it a go and report back,
Thanks for taking the time to answer my question.
Barry M

[BarryM]

Well, I gave it a go with these results.
First I took the relay out of circuit for another check on its function. Applying power to the coil several times, the contacts closed every time. Out of interest, I took my old AVO Mk 8 out of retirement and got a reading of 8 ohms with the contacts closed.
I added a 1 kohm resistor in series - the MP3 stayed silent.
I added a 100 nF capacitor across the contacts - the MP3 stayed silent.
I shorted the relay contacts - the MP3 burst into life.


Evidently there is something about the relay operation that is just not triggering the MP3. I have thought of a solid state relay but these all seem to be far too big to fit into the restricted space available. Have you any suggestions for what is going on and how to work around it?


Barry M

[HMS Invisible]

 You indicated a pressure switch so I naturally assumed you meant a momentary operation tactile switch that initiates a track playing. Can you confirm it was doing that and it is not a load carrying switch? Y/N
 
  A switch with trigger function  is usually gold plated to maintain minimum bounce when switching a signal to a processor at below one milliamp. It wouldn't be switching a significant current so just about all that could go wrong when extending wires and replacing with a working Silver or Cadmium plated >1 amp switch( or relay) is contact bounce confusing a processor input. The remedy for that is the typical 100nF and 1kohm.

It seems you just used a relay with a bad contact. The remedy is a better relay or you can use a bog standard opto-isolator for currents up to a few mA.
 Short of seeing the contacts and movement, current measurement outside expectation or contact voltage drop under load is a better diagnostic than a resistance measurement.
Out of interest, is this a reed relay or an rc product with relays?

[BarryM]

Thanks again for your quick reply and indeed your patience.
The answer to your first question is that it was just an initiator that triggered the MP3 to play. It was not a load carrier.
The relay used was a 6V similar to this [size=78%]Buy 6V 10A Relay | REUK.co.uk[/size].
I am happy to go down the opto-isolator route but faced with the selection such as here [size=78%]Optocouplers - Toshiba, Advanced Photonix and More | CPC[/size]  my brain freezes. Will you point out something suitable which I can use by the 'monkey see - monkey do' method?
Thanks again,
Barry M

[HMS Invisible]

I used Farnell & Digikey & have hundreds of optocouplers. But you will find the Ebay modules are useful for RC and time saving. You should be able to saw off a section of a multi-unit with screw terminals. Don't wreck a bandsaw blade cutting FR4 fibreglass by the way. Score & snap. You find optocouplers and relays in all sorts of E-waste.
Modules
https://www.ebay.co.uk/sch/i.html?kw=opto%20isolator
If you go down this route it would help to confirm the trigger switch current is less than one mA and successful trigger by ~ 470 ohm does that.
You then know a worst case value for I_t. Optocouplers have a current transfer ratio (CTR) of I_t/I_d and you see it in the device data sheet.
PC817 is a typical 50% so you bias the infra red diode for at least twice the minimum transistor current.
You will find a module's surface mount resistors bias for ~50mA so you should be able to dial it back with a higher value.

I'd be looking again why the relay didn't work, if it is really faulty, and if your eight ohm reading was really correct. Sparking damages the contacts, excess heat makes the soft plastic case melt and squeeze the soft insulation plastic. You could try levering off the case to clean or bend the contacts.

If you don't own a digital multimeter then I think it is worth you buying a £5 to £10 one.

Click thumbnail for full size.

[BarryM]

Thanks for that which gives me plenty to work through. I do have a digital multimeter and use it for most jobs but I'm old fashioned and, when the job gets serious, the old Mk 8 gets pulled out.
Barry M

[HMS Invisible]

 I'm about to order this https://www.ebay.co.uk/itm/255000634640 and other items from a UK located supplier. I'll post its circuit on Mayhem when I get the item.

Moving coil at ±2% FSD versus DMM at 1 part per million + a digit.
Now that I've had a chance to sleep on it I'll stick with what I have.