Death of my WD88VA XL - Page 5 - World-Designs-Forum

Shian7 · #41 17th October 2007, 03:43 PM

Quote:

Originally Posted by systemx

I do not have a WD88VA XL but how is the relay input panel connected?.

If you look at the board (in Nicks picture) with the diodes on theres a brown and blue connected to the 3rd and 4th tags (from top) on the right hand side, they go to the relay board. You can just see the edge of this board in the bottom right of the picture.

Mike.

Shian7 · #42 17th October 2007, 03:51 PM

I'm wondering if it's worth adding a fuse to all the outputs of the main transformer? If so what values? Are there any drawbacks to doing this?

Thoughts?

Cheers,
Mike.

systemx · #43 17th October 2007, 03:55 PM

Shian7.
Thank you, It just crossed my mind that there could be a problem, but the wiring is clearly o.k. On the KEL84, and I assume this amp, the screws which hold the panel in have to have washers to prevent the heater line from shorting. I was not/am not happy with this and replaced mine with nylon screws.

John T · #44 17th October 2007, 04:45 PM

Quote:

Originally Posted by Shian7

Thoughts?

Why?!

Bearing in mind the massive number of amps that have been built without them - with no problems .......

systemx · #45 17th October 2007, 04:51 PM

Yes there are problems with fuses. For them to be of any use. they tend to age and then fail for no apparent reason.
They can also affect the output of the amp as they start to act like resistors when they pass current near to there rating.

Shian7 · #46 17th October 2007, 06:21 PM

Awww sod it then!..... s'been fine for nearly a year now. Well, maybe I'll fuse the heater supplies. If I can be arsed! Sometime. Perhaps.
Having enough trouble finding time to even use it ATM, never mind modify it

Mike.

systemx · #47 17th October 2007, 06:47 PM

putting fuses in the heater line would be o.k. The tx will clearly take a bit of a hit from a short, so the fuse value will not be to critical.

NickW · #48 18th October 2007, 07:18 AM

All,

I think there may be something in the idea that I twisted the red wires too tightly - I will certainly be more thoughtful when I fix up the replacement.

All part of the training I guess !

I took a careful look at the underside of the tag board and there was nothing obvious in the way of a s/c.

I think the wires must just have got too hot !

Anyway, many thanks for all the help......

cheers, Nick

John Caswell · #49 18th October 2007, 08:21 AM

No, I don't for one minute think it is anything to do with the wires twisted tightly.
One of the rectifiers went s/c, current went up, the wires overheated, and the insulation melted as simple as that.
As for the comment by others that they have not known diodes "heal" themselves, yes it is certainly possible, as I know from a range of amps I repair. The heater volts disappear, then after a "cooling' period reappear, traced to a pair of 1N4007 in series to lower the volts slightly. Replacement by something physically larger cured the problem as it allowed more surface area for heat dissipation.

JohnT, bear in mind the massive number of amps that have been built with them, with no problems.

John - Dr John

Richard · #50 18th October 2007, 08:58 AM

Which diodes are used? No matter really, they don't look to be more than 3A, I can see 1N5 so perhaps 1N5408.

I earlier suggested 10A types mainly as it's easier, safer and cheaper (in the long run) to simply uprate well ahead of the "theory" but I'll expand a little.

The rating is supposed to be "average", ignore the one-off "surge" rating which doesn't refer to repetitive peaks. Here we are drawing a constant current so the charging time will be very short on peaks of the ac cycle. If they are only charging for, say, 10% of the time and the current out is 2A then they are conducting 20A whilst charging. Yes, the average may be less than 3A (we know it's 2A!) but that's not a lot of margin for a component in a hot environment. How well can they cool in the off periods during constant 100Hz peaks? I used 3A for Kit88 which drew a constant 0.6A.

Why 10A? Look at what's available and the prices. Many are moving to surface mount technology. Axials stop at 3A and the next best thing with legs on is TO220 package which are all similar size. Schottkies would be nice and quiet with low losses to allow a bit of voltage to be burned in the pi filter. Their surge rating tend not to be as good as regular types but we are going to rate generously. A quick look in RS at what is in stock shows many options. For example 50p each gets you 10A 45V TO220 Schottkies RS 486-0216.

All "up for discussion", what do you think?