Battery Charger problem

[JDNSW]

A couple of weeks ago, one of two battery chargers I have for backup of my solar system stopped working. Last time it did that, it was a fault with the control circuit board - $190, thank you.

Assuming it was a similar problem, I took it back to the people who repaired it a year ago. After about a week, they rang me and told me it was not repairable, as parts are not available.

A phone call to the manufacturer confirmed that there are two parts for this charger not available, the transformer and rectifier. And they were able to tell me that my repairer had enquired about a rectifier. They offered a discount price on their current model equivalent (actually only 35A not 40A) of just over $1000.

So I went back and got the charger; as a concession, they only charged me $50 instead of their normal $110 to look at it.

Today I pulled the cover off to see if I could find whether I could do anything with it.

Yes, the rectifier has clearly overheated, and a check with a meter shows at least one diode is U/S. The rectifier is a bridge rectifier, custom made with four discrete diodes individually soldered to two copper pipes as heat sinks. It seemed to me that I should be able to replace this, and a look at the Jaycar catalogue shows I can get a suitable bridge rectifier and heatsink for probably under $30 including postage. We will see how this works, but I can't see any reason why it will not be perfectly satisfactory!

It seems nobody is prepared to attempt this sort of modification today - took me all of five minutes with the catalogue to decide which parts are needed. Modifying it with the new rectifier should take no more than an hour. Even at commercial hourly rates this charger could have been repaired for well under half the cost of a new one. What is up with people?

John

[Graeme]

I'm reminded of when as a youngster I replaced a dead wafer rectifier pack in a Hornby train set controller with a single large diode and a big capacitor. The controller was being used for a slot car set that drew a little more current than the trains which caused the untimely demise of the rectifier pack.

[JDNSW]

Finished repairing the charger today. Total cost including postage and a couple of other bits (heat sink grease, some tags) and a spare rectifier - $43.

On close examination, the original setup was not a bridge rectifier as I thought, but a halfwave rectifier, or rather two halfwave rectifiers, using two paralleled diodes each. The transformer has two secondary windings, presumably because of problems winding it with heavier wire - the secondary windings are about 10g as it is.

I used two BR435 (MDA3504) rectifiers with the output paralleled, mounted on as large a fan-type alloy heatsink as I could fit in the space.

John

[Aaron IIA]

A repair facility couldn't adapt a new rectifier? Sounds like they just wanted to sell you a whole new unit.
Frequently the secondries on transformers are wired with a centre tap. This can then become the negative, with the positive coming through a single diode above the centre tap, and through a single diode below the centre tap. It could be seen as double half wave rectification.
Frequently I find discarded equipment that only needs a cheap fix. The most recent was a cordless drill that needed new brushes, and a lawnmower that needed a new sparkplug.
Aaron

[Kevin B]

I go to the once a year Jaycar returns sale, a truck load of returns dozens of piles of the same product, buy as is for a fixed price 4 or 5 of each product, I get them home on the bench and fix them up somtimes scrapping 2 to get one and the odd occation there has been nothing wrong at all, last one I went to I picked up 3 1000w inverters for a quarter the price of 1 and fixed all 3 for under 20 bucks,

[JDNSW]

.....
Frequently the secondries on transformers are wired with a centre tap. This can then become the negative, with the positive coming through a single diode above the centre tap, and through a single diode below the centre tap. It could be seen as double half wave rectification.
.....
Aaron

You could describe this setup that way - the two secondaries had four connections, but two were effectively joined as the negative, with the other ends of both windings going to two paralleled diodes each and thence to the positive.

John

[Pedro_The_Swift]

not knowing what I'm talking about at all,, just a quick observation,,
could you have installed bigger heatsinks if that was the original problem?

[JDNSW]

not knowing what I'm talking about at all,, just a quick observation,,
could you have installed bigger heatsinks if that was the original problem?

No, the original fault was one diode shorted, and the higher current flowing through its paralleled partner caused overheating.

A contributing factor may have been overloading - the charger, while rated at 40A, has a limit of ten minutes over 30A. Doesn't help that the built-in meter is rather wildly inaccurate (not surprising - it uses a length of 12mm copper wire as a shunt.

Adding a larger heatsink was not really feasible - the original design, before heatsinks were readily available, used two lengths of copper pipe as sinks, with the diodes soldered to them.

Any modern heatsink needed a complete redesign, which is effectively what I did.

John.