Repair cbus dimmers

In 2001, I had a new home built and had Cbus2 installed with three 8 channel DIN rail dimmers and one 12 channel relay. I also installed and integrated a Minder box and have been super satisfied with the flexibility and total integration this gave me. However, over the years, I have - apparently randomly - lost three dimmer channels - two on one dimmer and one on another. I have previously been able to reroute loads to keep things working, but the latest loss - apparently due to a new 240v halogen globe that seems to have destroyed the output - has left me beyond the point at which I can continue. To be honest, I really cannot possibly afford - or justify - paying what seems to be an obscene price to replace these two dimmers. I am a fully qualified electronics engineer with many years of experience, and I'd like to know if anyone has ever repaired one of these devices? I saw an old thread which suggested they were 'very difficult' to get "into" for 'regulatory reasons' (?) but is it possible? Does anyone have any diagrams? I also saw a thread that suggested the phase circuit was 'fairly simple' and I'm pretty confident that is where the problem lies. Blown triac, for example? There certainly appears to be nothing wrong with the CBus side of things. I really like my system, and have sung the praises of it to many, many friends - all of whom are terribly impressed! - but I just cannot see how I can justify spending at least a couple of thousand dollars to replace three channels. Does anyone have any comments; diagrams; or even practical tips to assist me? I'd be eternally grateful! (I have also been told that a repair facility is no longer available - and even if it were, what do I do with no lights for two or three weeks.....?).

Thanks - Mike

 

Can't specifically help you with your questions, and I apologise if my post irritates you, but I do notice that if you do have to replace dimmers, then if your bought one more unit, it would resolve your problems, leaving you with one channel spare.

 

A product repair service is available from Clipsal. You just take your faulty unit into a wholesaler who is selling C-Bus gear, fill out a Product Fault Report, and it gets sent to Clipsal in Adelaide for analysis. You then get a report stating either the repair cost or 'beyond economic repair'.

If you want to have a crack at fixing the unit yourself then I'd try and buy at least 1 other unit, so that you're not left without lights. C-Bus gear is on eBay all the time, so that might be worth a look if cost is an issue.

The 'regulatory reasons' probably relates to electrical safety. I think it effectively means that the teal/purple covers are glued on, however a link to the original post would be handy to see the context of that statement.

 

does the exchange service still exist.

a few years ago, my dimmer was faulty and it only cost a couple of hundred to give the old one back and get a replacement.

Mick

 

Exchange vs. repair is done at the discretion of Clipsal. Sometimes you'll get your exact unit returned repaired, other times you'll get a replacement of similar age. It depends on the nature of the fault, the amount of effort or type of repair, etc.

 

Repair dimmer unit

Thanks Darpa - that is a great idea (and it certainly doesn't irritate me! ). I'd have a "mechanical" issue - there is currently only space for four units on my panel - but that would be the least of my problems! A good example of lateral thinking....

 

When the latest problem occured, I called the CIS Help line in Adelaide and I was effectively told that '....Clipsal didn't offer either a repair or exchange service' and a new purchase was the only option. Has anyone had any repairs done recently - say in the last three months or so? Have they changed their policies?

 

I think the person you spoke to was mis-informed. Dimmers are definitely still repaired. I suggest you call a local electrical wholesaler, who sells C-Bus gear, and ask them about it.

 

Dimmer repair in the USA

I have a 4CH dimmer that has one bad output and since I am in the USA I am not sure if the Square D whom sells Clipsal will repair......
Anyone on the list knows or do I need to send across the pond.

 

CBus Dimmer Repair

OK, well that advice was taken and I picked up a - brand new - 8 channel dimmer from eBay. It was a good price - well, good compared to Clipsal!! - so now I can look at repairing the current one. However, it is a v1 and my units are all v2. I don't think this matters - I don't have any units except dimmers, relays, old version wall switches and, of course, my fabulous Minder box which makes it all so worthwhile.......

Should I be able to replace the faulty v2 dimmer with the "new" v1 and program via the Minder interface and the original CBus software accordingly? I guess I can but does anyone see any obstacles I might meet?

Thanks for all the - very welcome - advice!

Mike

 

I assume by this you mean it is a dimmer with teal coloured DIN Rail covers, and not the purple ones. The main change between C-Bus and C-Bus2 DIN rail units is the introduction of C-Bus Learn Mode, i.e. push-button programming.

You should still be able to program this unit using the Minder as the C-Bus Interface provided you use the C-Bus V2 Installation Software and not Toolkit.

The only thing to be mindful of is that the dimmer was under on-going development for quite a few years and plenty of running changes were made to make the dimmer more immune to things like mains ripple. If you get any weird or unexpected flickering or flashing, excluding mains fluctuations, it will be because of the older design of the unit.

 

Successful disassembly

Next instalment in the repair saga! I installed the 'new' ie replacement 5508D1A dimmer after programming it so that it looked like the faulty one. That worked well - the original CBus software is a breeze to use in this respect. The only issue is that one channel, which I am not using ie there is no load connected appears to be permanently 'on'. I have tried setting it to 0 on power up and also setting it to 'off' but the indicator comes on about 5 seconds after power is applied - presumably after the internal diagnostics have run. I can use the toggle switch to turn it 'off' but can anyone explain why a non-connected circuit does this? Is it a function of original CBus? (The replacement one has the teal co loured covers.)

I then started on the faulty dimmer to see if I could disassemble it. The unit consists of three sub units that are tied together with six metal plates and small screws on the rear. The sub units appear to consist of the CBus control on the left end as you look at the front, and then two identical units which appear to have four dimmer channels each. As I am convinced that it is a channel (ie 240v) problem, rather than a control (CBus) problem, I concentrated on the right hand sub-unit. It appears that the comment about the blue (teal) covers being glued on is correct. The cover has a side piece about 8mm wide, and this is where the glue is. Luckily, the glue is only minimal and I used a sharp knife to carefully cut the four glue 'spots' at each corner of the side of the cover. Each spot is at the top or bottom of the side, and you can use a sharp blade to carefully cut the glued pieces apart. Once this is done, there is a plastic clip in the middle of the front of each cover plate. Using a wide bladed screwdriver, you can carefully prise the cover off the (grey) back section. Once this was done, I could see that the glue was probably the same type used for plastic piping - and was indeed a very tiny application. Clearly, Clipsal don't want you to be "inside" without some effort, but if you are careful, it is not difficult. The gray case then splits in two parts - there are a couple of small plastic clips that are opposed to each other on the back. I had already removed the metal plates that tie this sub-unit to its neighbour, and also would stop the two parts of the sub-unit case coming apart. This exposes part of what appears to be the main channel PCB. The PCB actually spans the two channel sub-units and so I had to repeat the disassembly process on the middle one to get to the entire channel PCB.

With the second sub-unit case disassembled, the entire mains PCB is exposed. It plugs into the control sub-assembly with a four pin rigid plug and a small flat, flexible cable. This uses a small connector on the right end of the PCB and has a VERY small plastic retainer to hold it in place. Don't lose it!! The PCB has two groups of 4 circuits which are obviously the triacs and their respective snubber and control circuitry that switch the 240v supply. In addition, there is some control circuitry with what appears to be a memory chip at the base of the board. Each group of four triacs has a substantial heat sink attached. This just clips (pushes) on and is very easy to remove - just carefully pull it off!

At that point, I could see my problem! One of the triacs has obviusly gone s/c in a mildly spectacular way and slightly charred the board under it. The other that had failed also shows some evidence of damage, although it is much less. The triacs appear to be fairly standard - they are marked T1620600W and Googling this shows them to be a SGS-Thomson part T1620W 'snubberless triac'. See http://pdf1.alldatasheet.com/datasheet-pdf/view/25009/STMICROELECTRONICS/T1620.html for details.

At this point, I stopped. I still need to trace the circuit, but there is no evidence of any other component damage. The (presumed) triac switching device is a 6 pin DIL package marked 440H. This turns out to be a Low Input Current Optocoupler with a Phototriac Output - see http://www.vishay.com/docs/83631/83631.pdf for details.

I've added some photos and will post more as I find out.....

I'll post more as I work it out.....

 

This is interesting. - just a couple of comments:

1) The unused channel that is 'on', is likely to be a perfectly normal working channel that has had a level other than 0 stored for the power recovery option. Since the LED turns 'ON' 5 seconds after power is applied, the control part of the circuit is behaving as if this were the case. You should be able to make the unused channel power up OFF by simply storing a level of '0' in the power recovery option.

2) The components you have marked 'snubber' in your picture are actually EMC filtering components required to meet the requirements for CE marking (or C-Tick, for Australia & NZ). these parts are very robust and unlikely to have been damaged when the triac failed. You shouldn't have to mess with them.
Triacs can be damaged when incandescent lamp filaments break while the lamp is on, because an arc can form in the lamp, which bypasses large portions of the filament. The resulting current can be very high, well beyond the capability of most triacs. Chances are good that the triac is the only component damaged, although sometimes the damage can overheat tracks on the PCB.

 

Cost to repair

Hi All,

intresting, always wondered what they looked like on the inside.
just didnt want to take one apart that was'nt broken??
but one day....!!

As a matter of intrest what would be the cost for repairing this module through clipsal?

Aaron

 

Thanks, Don. I thought I'd done that but when I checked again, I obviously hadn't..... All is well now.

Yes, that all makes a lot of sense. I hadn't really traced out any circuitry when I posted the pictures so it was a guess.

I think you have hit the nail with the triac problem. The latest failure was almost immediately after replacing a GU10 240v halogen globe. I suspected something like you suggest. There is quite clear evidence of one triac that has suffered a major overheating with scorching of the PCB - although the tracks look good. I will have to see if the damage is severe when I take the triac off the board. The other - faulty - channel also shows signs of overheating, although it is much less catastrophic. My only fear, I guess is that some of the associated circuitry, much of which is surface mount, has been damaged. As you say, though, and as I have seen in the past, with "luck", it will only be the triac that failed.

Thanks to everyone for your very helpful comments - and I guess I'm glad that it isn't a newer device if they are far harder to get into!

 

Hi Mike great photo's especially the one showing the modular construction of the older style output unit.
I once asked the question at the UK head office why a combination unit was not developed from this modular construction say four dimmable and four standard relay channels, obviously with appropriate GUI. Still don't know why they didn't develop my idea

I have experienced lost channels also on the dimmable output units, once from a standard filament candle lamp pointing upwards and once from a GU10 downlight pointing downwards. Is there a preffered install method ?? too protect these vunerable Triacs.

 

If you have complete control over the construction of the luminaire, you might reduce the number of failures by mounting incandescent lamps so that if a section of the filament drops, it falls away from any supporting wires, rather than back into the supports. This way, the chances of new arc being established which bypass the filament are reduced. No guarantees though.
Use of 12V lighting with transformers can help here, as the filament arcing is less of a problem at lower voltages, and the series impedance of the transformer further limits the peak of any resulting current surge.

The problem of triac (and copper track) damage when incandescent lamps 'blow', has been around for a long time. When the event occurs, very high currents are possible, and all elements of the circuit from the switchboard, to the triac, the wires, and the lamp are subjected to the same current surge.

In many installations, the peak surge current is limited by all the wires, connections and transformers in the path back to the power station, and triac failures are rare. In other installations, the peak surge currents can be very high, and triacs can fail regularly.

As all elements have some voltage drop or resistance associated, there is a power loss associated with the high current which will raise the temperature of each part of the circuit. All materials in the circuit will fail at some point because of the power dissipation, and the trick to protecting the triac is to ensure that something else fails (or breaks the circuit) before it fails.
All triacs have an I-squared-t rating, to help coordinate protection. For the triac in the dimmer, the I-squared-t is 190 Amp-squared-seconds. (This is pretty high for a 16A triac, btw, and higher values usually mean larger, more expensive components). The easiest way to protect this would be to install a series fuse with a lower 'clearing I-squared-t'. I-squared-t limiting fuses have some construction feature (sand filled, notched fuse element, etc.) to break any arcing into small sections so that the arcing can be quenched effectively. These fuses are often called 'electronic' types, and can be expensive.
It is also possible to protect the triac with a fast acting circuit breaker with an arc extinction chamber.
The hardest part of this coordination of protection is finding the I-squared-t figures for fuses and circuit breakers.
Installing fuses also comes at a cost, and to be useful, the fuses would have to be mounted so that they can be accessed. Fuses with a lower I-squared-t are likely to fail more often than the triac, and this can become a new problem.

I hope I've answered your question. There is no easy answer.