**FIXED** Shutter circuit fried...replacement or repair?

[Paul M]

12 hours ago, symmetal said:

An arduino nano could easily be programmed to replicate this if the function of the CL and CP signals were known.

To me it looks like "OP", so maybe they are the switch lines from the Open and Close limit switches.

I'm baffled why a PIC would be selected for the application at hand!

[steppenwolf]

19 minutes ago, Paul M said:

To me it looks like "OP", so maybe they are the switch lines from the Open and Close limit switches.

I'm baffled why a PIC would be selected for the application at hand!

These are indeed the Open (OP) and Close (CL) limit switches

On the other connector, OP1 and CL1 are connected to the manual Open/Close switch, OP2 and CL2 are connected to the NO connectors on the RF relay board.

With regard to the PIC solution, I agree because full control can by organised using three relays and four diodes (excluding the flyback diodes for the relays)!!

[tooth_dr]

@symmetal Alan thank you so much for this diagram, quite impressive 

 

So if I understand what has happened here:

1) initially shutter stopped working due to 'unknown' reasons

2) my 'testing' caused a short and burnt out a track on the board, which may or may not have damaged relays etc.

3) subsequent replacement of capacitors and relays has made no difference, and the fault was eventually traced to a failed voltage regulator

4) the voltage regulator failed allowing 12v through to the microchip, and subsequently destroyed it - which explains why it initially stopped working

5) replacement of voltage regulator is pointless due to the damage that has been done already to the PIC microchip?

 

Options

1) find a replacement PIC microchip - but I cant just buy one of these because it needs programmed, is that right?

2) build another circuit to control the shutter - @steppenwolf I think this is where your board comes in?

 

I'm trying to work out what is going on overall with the shutter and dome rotation system - have i got this right?

                      --The CEBEK TL-2 board, a readily available manufactured board, receives a bluetooth signal from the dome rotation controller system (in my case Shelyak) activated via POTH on the PC.  Depending on whether I click open or close shutter on POTH, a different part of the CEBEK TL-2 circuit activates and this send a voltage via a wire to the failed board (the board that has failed).  The failed board is solely concerned with controlling the shutter motor, and simply receives a trigger via the CEBEK TL-2 board.  The failed board also appears to be able to control the shutter via a rocker switch - so this would be another trigger - albeit a manual one.  Would I be able to implement this manual control feature as well if I make another board (or is that even necessary - but possible useful in the event of a power failure or PC fault and needing to close the roof).

 

 

Thanks for all the input guys, really really incredible.

 

 

 

[DaveL59]

I can't give a definitive answer to your options but I believe a new PIC would need to be programmed.

For the what happened, sounds about right. It's very likely the PIC got fried when the regulator failed, tho it's a simple job to replace it and roll the dice and see if you got lucky. If not it can be removed and put in the spares drawer 😉 I guess the 7805 wasn't heatsinked but then it's a pretty low power circuit so may have been fine like that.

Sounds like the bluetooth module switches the open/close command via relays, effectively being the same as pressing a manual button so the simple circuit steppenwolf describes may well suffice if the bluetooth module holds the relay closed for a long enough duration or the motor controller will operate on a momentary trigger.

 

edit - having said this, I realised that the relays you replaced are 12v rather than 5v, so I'm not so sure you'd be able to prove function if you did swap out the 7805. Possible your original relays may still be functional so if you swapped those back onto the board as well you might be able to determine if there's life...

Edited October 4, 2020 by DaveL59

[symmetal]

Glad to help Adam, I like a challenge. 😀

Yes, it's bound to be OP and CL signals coming into the board. I first read it as CP and didn't look at it closely again. 🤭

Your final analysis is correct Adam. 780x series regulators are pretty robust with thermal overload and short circuit protection so it's unlucky it failed.

A replacement board is the best solution, I think, based on steppenwolf's design looks easiest, as the PIC solution is not on if you don't have the programming facilities. An arduino solution is a bit of overkill for what seems a fairly simple requirement. Once you and steppenwolf have worked out a working design, If you wish, I can do a board layout using KiCad which can then be sent off to have it professionally made. Or, for a one-off, I can do a less professional, home made one. 🙂

Alan

[steppenwolf]

53 minutes ago, tooth_dr said:

--The CEBEK TL-2 board, a readily available manufactured board, receives a bluetooth signal from the dome rotation controller system (in my case Shelyak) activated via POTH on the PC.  Depending on whether I click open or close shutter on POTH, a different part of the CEBEK TL-2 circuit activates and this send a voltage via a wire to the failed board (the board that has failed). 

Yes, this.

53 minutes ago, tooth_dr said:

The failed board is solely concerned with controlling the shutter motor, and simply receives a trigger via the CEBEK TL-2 board. 

Yes and is the connection for the manual switch.

54 minutes ago, tooth_dr said:

The failed board also appears to be able to control the shutter via a rocker switch - so this would be another trigger - albeit a manual one. 

Yes

1 hour ago, tooth_dr said:

Would I be able to implement this manual control feature as well if I make another board (or is that even necessary - but possible useful in the event of a power failure or PC fault and needing to close the roof).

Yes

[Davey-T]

Did a bit of delving into PIC programming and it ain't simple, needs a lot of software and some programming experience so definitely not worth the hassle 😆

Dave

[symmetal]

8 minutes ago, Davey-T said:

Did a bit of delving into PIC programming and it ain't simple, needs a lot of software and some programming experience so definitely not worth the hassle 😆

Dave

You also need the programming hardware with 3 power rails to program the EEPROM in the PIC. 😀 Datasheet. 

Alan

[Davey-T]

1 minute ago, symmetal said:

You also need the programming hardware with 3 power rails to program the EEPROM in the PIC. 😀 Datasheet. 

Alan

Yes saw that, guess it's ok for mass produced stuff, Lakeside motor focuser is PIC controlled,  but not worth it for homework, maybe they've been superseded by Arduino , RP for that.

Dave

[symmetal]

4 minutes ago, Davey-T said:

Yes saw that, guess it's ok for mass produced stuff, Lakeside motor focuser is PIC controlled,  but not worth it for homework, maybe they've been superseded by Arduino , RP for that.

Dave

Yes, Pics were the bee's knees 25 years ago or so, but are rather old-school nowadays.

Alan

[steppenwolf]

1 hour ago, symmetal said:

A replacement board is the best solution, I think, based on steppenwolf's design looks easiest, as the PIC solution is not on if you don't have the programming facilities. An arduino solution is a bit of overkill for what seems a fairly simple requirement. Once you and steppenwolf have worked out a working design, If you wish, I can do a board layout using KiCad which can then be sent off to have it professionally made. Or, for a one-off, I can do a less professional, home made one. 🙂

Alan

Hi Alan, part of my circuit diagram is taken from this online resource which was intended for use with the Lesvedome observatory control system. However, the two boards (RF board and relay switching board - the one that was fried) are identical it would seem in both the Shelyak system and the Lesvedome system that I had. It might make more sense if you have a look at the link to see what is going on with the diodes, the single relay (RY1 on my diagram) and the limit switches which all pertain to the left hand part of the circuit diagram that I posted earlier in this thread. The two additional relays (RY2 and RY3) swap around the polarity to the motor depending on what happens at the relay switching on the output of the receiver module and also allow for manual operation of the shutter using Auto/Manual Selection switch (S3) and the existing Open/Close switch built into the control panel (S4). Switch S5 simply turns the whole kit and caboodle on and my guess would be that this switch is present on the front panel alongside S4 but perhaps @tooth_dr can confirm its presence.

[symmetal]

2 hours ago, DaveL59 said:

Edit - having said this, I realised that the relays you replaced are 12v rather than 5v, so I'm not so sure you'd be able to prove function if you did swap out the 7805. Possible your original relays may still be functional so if you swapped those back onto the board as well you might be able to determine if there's life...

To save swapping the relays back you could make a slight modification to the PCB by cutting a track and adding a link which would power the relays from 12V instead of 5V.

Or just monitor the PIC outputs on pins 6 and 7, that they change from 0 to 5V when the relays are meant to operate, which would indicate that the PIC somehow survived.

Alan

Edited October 4, 2020 by symmetal

[tooth_dr]

So.... I soldered back in the original relays and fitted a new voltage regulator which I picked up this afternoon off a friend.

lol and behold it’s alive 🤷

I honestly can’t believe it.

I have to say a massive thanks to everyone in this thread who provided input, insight and solutions.  Without this help I would genuinely still be manually pulling the shutter up and down balancing on a wobbly stool.

I have a feeling that I may end up back in here again if the system fails again but at least I now have some idea of where to go and hopefully this thread proves useful anyone else with similar problems.

 

Adam. 

[DaveL59]

Wow that is good news Adam, surprised it survived with 12v hitting the 5v rail but great result all the same, plus you know some options if it barfs again to troubleshoot or build a replacement 🙂 

Out of interest, how warm does that 7805 run? Might be worth fitting a heatsink just to improve its longevity.

[tooth_dr]

17 minutes ago, DaveL59 said:

Wow that is good news Adam, surprised it survived with 12v hitting the 5v rail but great result all the same, plus you know some options if it barfs again to troubleshoot or build a replacement 🙂 

Out of interest, how warm does that 7805 run? Might be worth fitting a heatsink just to improve its longevity.

It got good repeated doses of 12v too 😂

What sort of heat sink do you suggest?  I’m sure I can fit it ok anyway as a preventive measure. 

[DaveL59]

yeah I expect it did, had to be worth a few mins and a low cost part to see and it paid dividends 🙂

For a heatsink, something like this should do fine

or if its getting very warm (hot) then something like this

Just be careful when fitting that it isn't in contact with any pins, and be aware that that big metal tab is connected to one of the pins (centre=ground) so any short from the heatsink to another contact point or component risks blowing things again.

 

Edited October 4, 2020 by DaveL59

[tooth_dr]

14 minutes ago, DaveL59 said:

Just be careful when fitting that it isn't in contact with any pins, and be aware that that big metal tab is connected to one of the pins

Thanks again. Didn’t realise that, will be careful 👍🏻

[steppenwolf]

What a fantastic result and VERY pleasant surprise - Hooray!!!!!

[symmetal]

That's good news Adam. 😃 I didn't expect that.

It may be because the PIC chips are actually CMOS based and not TTL as I first assumed, although they use a 5V supply. CMOS is more tolerant of voltage limits than TTL. The PIC power pins are labelled Vdd and Vss which should have been a clue. If it was TTL they would be labelled Vcc and Gnd. 😳

Alan

[tooth_dr]

1 hour ago, steppenwolf said:

What a fantastic result and VERY pleasant surprise - Hooray!!!!!

Thanks for your input Steve and offer of help. 

31 minutes ago, symmetal said:

That's good news Adam. 😃 I didn't expect that.

It may be because the PIC chips are actually CMOS based and not TTL as I first assumed, although they use a 5V supply. CMOS is more tolerant of voltage limits than TTL. The PIC power pins are labelled Vdd and Vss which should have been a clue. If it was TTL they would be labelled Vcc and Gnd. 😳

Alan

Thanks Alan.   I’m glad that it’s survived the extra voltage, but what you say above explains why it did. 

[Skipper Billy]

I can't pretend to understand 90% of the technical aspect of this thread but I do recognise several rare and special things.

People with specialist knowledge that are happy to share it freely

People helping each other because they can

People seeing an issue through to its end

Men asking for help!!! 😉 

Well done all of you.

As an observer this is truly heartwarming.

 

[DaveL59]

54 minutes ago, Skipper Billy said:

I can't pretend to understand 90% of the technical aspect of this thread but I do recognise several rare and special things.

People with specialist knowledge that are happy to share it freely

People helping each other because they can

People seeing an issue through to its end

Men asking for help!!! 😉 

Well done all of you.

As an observer this is truly heartwarming.

 

talking of which, did you manage to get the mirror on your 5d sorted in the end?

[wulfrun]

A bit late to this topic but if the voltage-regulated circuitry is sensitve to over-voltage, it ought to have a "crowbar" zener diode following the voltage regulator. A 5.6V 1-watt zener to earth would cause the fuse to blow if there's an over-voltage fault on the 5V line. Any series-regulator that relies on a bipolar pass-transistor is liable to fault by short-circuiting and passing full voltage, since that's almost always the failure mode of the pass-transistor. Commercial designers don't bother with such protection unless it's costly circuitry being protected, since it costs in components and PCB space - although in this instance they did bother wth reverse-polarity protection from the diode at the power input, strangely.

[tooth_dr]

1 hour ago, wulfrun said:

A bit late to this topic but if the voltage-regulated circuitry is sensitve to over-voltage, it ought to have a "crowbar" zener diode following the voltage regulator. A 5.6V 1-watt zener to earth would cause the fuse to blow if there's an over-voltage fault on the 5V line. Any series-regulator that relies on a bipolar pass-transistor is liable to fault by short-circuiting and passing full voltage, since that's almost always the failure mode of the pass-transistor. Commercial designers don't bother with such protection unless it's costly circuitry being protected, since it costs in components and PCB space - although in this instance they did bother wth reverse-polarity protection from the diode at the power input, strangely.

Thanks for the reply.  If this is something I can do, then it would make sense to do it.  Can you explain how I would be able to add this to the circuitry?

[wulfrun]

1 hour ago, tooth_dr said:

Thanks for the reply.  If this is something I can do, then it would make sense to do it.  Can you explain how I would be able to add this to the circuitry?

Easy - obtain a Zener diode of 5.6V and 1-watt (or higher) rating. Referring to the hand-drawn schematic further up this thread, connect the zener between the positive side of C3 and earth, the other side of C3 will do nicely. The Zener has a band on it, that end must be connected to the positive of C3. Use a reasonable gauge wire or just solder the Zener on the back of the board, whatever is convenient physically really. Zeners are available in a wide range of voltages, you need something higher than the supply is ever likely to be in normal use but not too much higher. Off the top of my head, 5.6V is a standard value and quite acceptable. Be aware that if the zener is called into action, it might be sacrificial and fail itself but it will certainly blow the fuse and protect the other circuitry. Just to explain, a zener that is wired "backwards" passes almost zero current until the threshold voltage is reached, above which is rises very steeply. It will draw so much current that the fuse will blow, and if it fails in the process it will fail short circuit too, guaranteeing protection. Excellent and cheap over-voltage protection known as a crowbar circuit.