Victron Charger Poorly or Dead ?

[cuthound]

Electrolytic capacitors often dry out if not used for a while, then fail when  energised.

[WotEver]

Just now, cuthound said:

Electrolytic capacitors often dry out if not used for a while, then fail when  energised.

Yup, they're generally the first thing to change on anything over 5 years old, particularly something that gets hot like a charger. 

[mross]

OP has ordered a new one.

[WotEver]

2 minutes ago, mross said:

OP has ordered a new one.

Charger?

[mross]

Yes, I got a PM.

[WotEver]

I'd be happy to take a look at the old one if a schematic can be found. 

[Alan de Enfield]

9 minutes ago, mross said:

OP has ordered a new one.

A 'wee bit early' - but - yes that is the plan for Monday (as the dealer is closed at the weekend)

Having found an identical (but hopefully un-dead) model at under £400, (new with 5 year guarantee) I think that is the 'way to go'. If the old one turns out to be repairable, I can keep it as a spare, or swap it out on the other boat which only has a 40A charger.

I have not totally given up, and will continue to 'poke about' with this one and post a couple of pictures of the board.

I may also give the repair man a call 1st thing Monday to see if it is possible to get an estimate.

[Alan de Enfield]

5 hours ago, Giant said:

So first question: is it still blowing fuses? You've replaced the input fuse and powered it up, now take that new fuse out again and check if it's been blown.

If the new input fuse has blown, then something is still more-or-less shorting out, and the strategy is to find that low-resistance path.

If it hasn't, then something did fail and briefly drew a high current in the process, but has now burned itself out such that it no longer shorts -- but presumably no longer functions either. The strategy then is to identify what blew, either by visual signs of failure (which you've drawn a blank on so far) or by detecting a lack of function.

Whatever it is, I would expect it to be within the mains side of the circuit, because a fault on the low voltage side would have to pull at least 3.6kW through the transformer to blow the mains input fuse and you'd expect something else to have obviously gone bang in that case.

The most obvious candidate would be an electrolytic capacitor. Often the first things to fail in any electronics, but particularly common on the ones that are exposed to mains voltage, especially if ever exposed to spiky supplies.

Look for any signs of bulging in the ends of capacitors.

Input Fuse - has continuity

Output Fuses - have continuity

Capacitors (round drum like things) show no sign of distortion, smell, or 'blacking'

What is confusing me is that I am getting 11.25v at the output terminals with the on/off switch in either position

 

Pictures :

1st Picture 'main'  input bottom RH side, with 'output' terminals running up the RH side

The four 'solder blobs' to the left of the 'mains' input are the location of the 15A input fuse (on the reverse of the board)

 

Edited March 12, 2017 by Alan de Enfield

[mross]

Where has all that 'toothpaste' come from?

Edited March 12, 2017 by mross

[Chewbacka]

1 hour ago, mross said:

Where has all that 'toothpaste' come from?

Looks like silicone rubber used to support the components must susceptible to cracked 'legs' due to vibration. 

[Alan de Enfield]

5 minutes ago, Chewbacka said:

Looks like silicone rubber used to support the components must susceptible to cracked 'legs' due to vibration. 

Yup - its 'mastic' 'wubbery' type stuff.

[Debby]

From Dave via email as he is upstairs in his "music room"

Interesting thread, I've only glanced at the boards but can I add a few things which may be useful ?  Don't know if any of these would help. A schematic is really the first thing to refer to. 

General tips, then. 

On the control PCBs, check the voltage regulators, if there are any.

It might not look like this, some PCBs use "voltage references" look more like a conventional transistor package. 

 

The thread is correct in saying that electrolytic capacitors should be replaced, but voltage regulators tend to be overlooked. There's a line of 4 transistor like ICs on the board and two transistors to the right..... check to see if they have standard transistor markings on, such as BC547 or 2N3906. 

There are some fairly large diodes on the board (1N40XX ?) which are presumably part of the rectification. Check these. 

If the electrolytic caps have been replaced, check the polarity. They're polar components, white stripe on the case is the negative side. 

I'd check voltages for the chips. Get the pinouts from a datasheet and see if the correct voltage is there for positive and negative feeds.  

PS Checked and there IS a voltage regulator like the above one. It's near a 4 x 4 matrix of solder joints, with two flat yellow capacitors (R140 / R147) next to it. 

There will probably be two voltage regulators in there, sending +9v/-9v (or 12v) to the electronics.

These go on a regular basis. They're the usual suspects with dead PC power supplies.

Apologies if non of this is appropriate, my experience is in microelectronics. 

 

Edited March 12, 2017 by DaveandDebby
To add more

[Debby]

There's more...

Aha.

Two resistors - are these shorted ? There's some obvious Friday afternoon soldering here. 


D12 above it seems to be two 1N4148 diodes which have been modified. Presumably a higher rating diode was needed, so it's been modified by adding two 4148's together. What the lead is which comes off the diodes is, or where it goes, or whether or not it comes off them at all, is an interesting question. 

PS Surface mount fuse.... I doubt it, depends how much current is going through a fuse, if that's the culprit.

Surface mount is easy to repair. You need to heat the component, remove it with tweezers, clean the pads, apply solder paste, add a new component and then seat and re-heat the component. However, a cursory glance at the board doesn't reveal any surface mount components. So unless it's hidden on the back of a board..... I think it's not going to be down to a SMD component. 

 

Edited March 12, 2017 by DaveandDebby
More stuff I don't understand!

[WotEver]

1 hour ago, DaveandDebby said:

D12 above it seems to be two 1N4148 diodes which have been modified. Presumably a higher rating diode was needed, so it's been modified by adding two 4148's together.

My thinking is that they've modified D12 to be two diodes in series in order to give double the voltage drop. IN4148s tend to have a lower voltage drop than say a 15920 so doubling them up is a pretty common bodge. 

Tony

[Giant]

4 hours ago, WotEver said:

I'd be happy to take a look at the old one if a schematic can be found. 

You won't find one. Unfortunately the good old days are long gone when the schematic was to be found in the back of the manual because it was expected someone might need to repair the product.

These days schematics are almost universally closely guarded secrets, shared only with approved repairers if at all.

The only way to approach DIY repairs now is usually to work through the PCB trying to identify the functions of things, figuring out what voltages and waveforms should be expected where, and trying to find the first point in the system where things look wrong. When things get complicated, sometimes even drawing out parts of the schematic by tracing out the connections visible on the PCB.

In this case, with a two-layer board and lots of easily identifiable through-hole components, that's reasonably doable - but it's still slow work. And hard to do without having a good knowledge of common design patterns and faults, the board powered up in front of you, and an oscilloscope to see what's going on. For multilayer boards covered in unidentifiable SMT components, it can be nigh on impossible.

Looking at this one, I'm afraid I don't really have any more obvious ideas for Alan to try. I'd need it sitting in front of me to try and probe through it and understand what's going on.

[Debby]

Hi WotEver, Dave here. Yep, could be. 

I think it's equally likely that some lazy gentleman somewhere couldn't lay his hands on a proper 1N40xx diode, so he just joined to 1N4148's in series, which is pretty ...... poor.

For a £600 bit of kit, I expect to see things like IC sockets. Pennies each. As little point wiring as possible. Proper heatsinks added to voltage regulators. Which produce heat and therefore can suffer from damage from inadequate heat dissipation (which is another reason which is why I think this is a voltage regulator issue). I also expect to see properly sleeved joints using heat sink and not any old bit of plastic tubing which was lying around and a general lack of insulating foam would also be nice.

As well as voltage regulators, there are also voltage references. Pic of one below. These basically contain a zener diode across legs 2 and 3. Zener diode produces a regular voltage. Regulators and references are what you generally use if you want part of a circuit at a known voltage, say, 10v. Unfortunately, they're a bit more difficult to distinguish by newbies from transistors. 

When testing regulators and references, it's pretty simple. Multimeter out and look for a voltage difference on either side of the component. 

A word to the wise.

Before replacing a faulty one, be CERTAIN of what type of regulator it is. For example, a 78L10 produces PLUS ten volts. A 79L10 produces MINUS ten volts. This will result in electrical unpleasantness and lots of white smoke, which is not the desired effect. Luckily you can't find 79L10's for love or money, if you do, send them my way, I have some voltage controlled oscillators to build for a synthesizer. ;-)

 

[Debby]

 

Edited March 12, 2017 by DaveandDebby
Duplicate post

[Debby]

PS Good post from Giant above and 100% correct. ;-)

This is the kind of stuff I do. Note the horrifyingly large components and 20x assembly microscope. ;-)

[Debby]

Whoops, apologies for coming back in all of a sudden.

Just noticed. Shot with the two suspect resistors and diodes above ? I can see two tantalum capacitors in there as well.

Tantalum caps are notorious for giving up the ghost if polarity to them is reversed even once. So there's one to check. Could be 1001 things, but the top of my list is..... check the voltage regulators and references on the boards. 

[Alan de Enfield]

Thanks for the comments Gents - way above my head - wiring OK, but ''black centipedes with bits sticking out - no way.

 

I am not sure about the 'DIY' and 'they have done' comments as it certainly hasn't been touched (or even opened) whilst in my ownership. Maybe Victron is not the 'quality' produced we (I) think they are.

Jobs for Monday :

1) Phone Victron repair man for an estimate

2) Contact Victron dealer that has one for sale at £200 below RRP.

[Debby]

Hi Alan, sorry for getting electrically confusing there ! 

Hopefully you'll have some luck with Victron.

Yes, the board is probably untouched and that's how they are manufactured. Apologies for pointing it out..... it's a bit rough and ready in there in places. 

All the best with it, hope there's a good, and cheap, result. 

[Giant]

50 minutes ago, DaveandDebby said:

For a £600 bit of kit, I expect to see things like IC sockets. Pennies each.

I wouldn't want to see IC sockets in a marine environment. You want a solid connection that can't get separated by corrosion.

[Debby]

33 minutes ago, Alan de Enfield said:

Thanks for the comments Gents - way above my head - wiring OK, but ''black centipedes with bits sticking out - no way.

 

Sorry, Alan. This is why I no longer say "What have you been up to in the workshop this weekend, dear?" to Dave. Three hours later I'm no wiser, but I have a headache.

 

Debby

[Alan de Enfield]

Any thoughts on the Electroquest range of chargers ?

I can get 1x 100a (twin output) or 2x 45a (twin output) for the same price as the Victron

https://electroquestuk.com/chargers/leisure-battery-chargers-marine-battery-chargers/fully-automatic-4-stage-marine-industrial-battery-charger-12v-100a

 

I have 4x 230Ah FLA 'domestics' and 2x 230Ah FLA 'starter' batteries, so if I go the route of 2 x 45A chargers then one can power up the domestic bank and the other, one-output to each of the engine batteries.

This also gives me a 'fall-back' charger should either of them fail - cunning plan ?

 

 

[Giant]

7 minutes ago, Alan de Enfield said:

Any thoughts on the Electroquest range of chargers ?

I can get 1x 100a (twin output) or 2x 45a (twin output) for the same price as the Victron

https://electroquestuk.com/chargers/leisure-battery-chargers-marine-battery-chargers/fully-automatic-4-stage-marine-industrial-battery-charger-12v-100a

Electroquest sell a lot of different chargers from a range of manufacturers. The ones they sell that look like that, with a red casing and black front panel with red LED display and fans, are made by a Taiwanese company called Fairstone - http://www.fairstone.com.tw/english/index.htm

I don't see any evidence of testing & certification for them anywhere, and it seems the manual comprises these two pairs of pages. I don't think I would spend that much money on one, whatever its claimed specifications.