Oh b*gger, I killed it.

Yes that's right I've broken something.

A little while ago I picked up an Ariston AX-910 integrated amplifier at a car boot sale for just £4.

The AX-910 is essentially a re-badged Cambridge Audio A1 mk3 but with a simplified pre-amp board.

I have downloaded the schematic for the A1 mk3 and the power amp boards are identical.

To cut a long story short, this evening I forgot to turn the power off whilst 'messing about ' inside the amp it went BANG. Lesson learned.

I'm pretty sure that what happened is that I caused a short circuit between points 'A' and 'B' marked on the image below.



The short resulted in the expected 'POP' and sparks. I turned the amp off as quickly as I could and found that fuse F3 had blown (T 4A).

Initially I thought/hoped that maybe the fuse blowing had prevented any further damage but after I replaced the fuse and turned the power on again one of the TDA1514A IC's (IC U8) went up in smoke.

I have already ordered a pair of replacement TDA1514A IC's.

When these arrive what procedure should I follow and what checks should I make before powering up the amp again? I guessing that simply installing the new IC's and 'hoping for the best' is not the best idea.

 

Ah yes, the flash and crack of doom from the slip of the screwdriver or probe!

As you've shorted the positive to negative rail it might well have buggered the rectifier diode bridge.

Remove the power op amps and measure the DC lines.
If the fuse still blows with no op amps in circuit then the diodes definitely need replacing. If you get the DC rails as indicated, or a little higher as they are unloaded, a new TDA should have it working.

It's possible the PSU caps are damaged if the diodes have failed 'short' but likely not 'fatally' so to speak given you powered off rapidly and the fuse blew. I guess you'll replace them anyway as its an old amp.

If you haven't already tried this, get yourself a set of spare probes and insulate the ends with a few turns of tape leaving just a mm or two at the tip.
A lifesaver for those inevitable little slips:


IMG_0872 by trebor1966, on Flickr

 

If it is tricky to remove the whole (working) TDA1514 you can just lift / snip the two power pins (pin 6 and 4). That power section schematic looks simple so as Rob suggests if the PSU lines come up well (without the TDA1514 connected) it will probably be okay when you put both back in.

All caps are probably okay. Maybe worth looking at the ripple on the lines to see if it is excessive. A 'scope would be best, but you could use a multi-meter set to AC. I'm not sure what would be a standard amount to expect though, lol. I suppose both + and - lines would be similar if all is okay.

Of course the pre-amp section might be damaged too?

Check for damaged PCB traces as the large surge of current can melt them.

 

Thanks Rob and Simon,

There's something incredibly disappointing about the smell of toasted IC's.

There are two wires which take the +/- 30v (ish) DC from the main power supply on the power-amp board to the pre-amp board and it's these two wires which were allowed to short against each other.

These two wires were disconnected at the pre-amp end at the time of the 'accident' meaning there was no power to the pre-amp board. The pre-amp section should therefore be OK any damage should be restricted to the power-amp board.

Unfortunately I had just finished refurbishing the amp with new capacitors etc so hopefully the brand new caps are not dead.


I was up early this morning and this is what I have found so far:

I have removed all four rectifier diodes and they all test correctly with my multimeter in that they show continuity in one direction but not the other. New diodes are cheap enough so I'm happy to replace them as a precaution if you think it's wise.

I have removed all of the electrolytics they all test within spec for ESR and capacitance. I've no idea how to go about testing ripple and I don't have a scope.

There is no continuity between the transformer secondaries or between the primary and secondary windings and the voltages on the transformer secondaries are correct.

The TDA1414A for the left channel is definitely toast and has left light scorch marks on the PCB. I have removed this one already.

The other TDA chip (right channel) doesn't show any visible signs of damage. Is there a way to test this in any meaningful way with a multimeter or should I just fit a new one to be sure? I've already order two anyway.

My capacitor tester doesn't go lower than 0.1uF so I haven't tested any of the small value capacitors.

I haven't tested any of the resistors at all. Should I do this?

So I'm thinking that I should now refit all of the capacitors and the rectifier diodes and with both TDA chips removed power up the board and test the DC voltages after the rectifier and also at the power pins of the TDA's (pin 6 and 4). Does that sound sensible?

 

Oh and thanks for the tip about the probes Rob.

 

Sounds like a sensible plan. If you get the 30v rails and no fuses blow you can safely install the new chips.

I'd only start testing the resistors and small caps if they show signs of damage.

If the capacitance and ESR of the main caps is measuring within spec those will also be fine.

 

Excellent, it sounds like I may have got off fairly lightly then.

I'll report back once the new TDA's have arrived.

As a side note, the PCB has the facility to accept either 8 smaller power supply filter capacitors or alternatively 2 large snap-in type can capacitors.

The amp came fitted with 8x 2200uF cheapo capacitors and I had swapped these for 8x 3300uF Panasonic NHG.

I also have in my parts bin some used but still perfectly serviceable 10,000uF Elna snap-ins which came out of my Linn Klout.

Is there likely to be any difference either way in terms of sound quality?

 

No, I'd leave the NHGs in place.

 

Cheers

 

I think personally I would keep the original right channel chip in when trying to power up as 'fixed'. If the PSU lines are okay and the IC is broken it seems unlikely it will damage anything. If it does work you have a spare IC.

On the other hand if something else is still wrong, and you put both new ICs in, you could blow them both.

 

Thanks for all the great advice, unfortunately I've just realised that you need to fit the IC's before fitting 12 of the electrolytic capacitors which I have already removed to test them. Doh!

This is because once the capacitors are installed they block the access to the two screws which are used to tighten the IC up against the heat sink.

I think I'm going to have to fit all the caps, test the voltages, remove the caps again to fit the IC's and then refit the caps again that were in the way.

That's quite a lot of work so I might just fit both of the new IC's straight away to save me the hassle of having to repeat all that if the old IC from the right channel does go pop.



With any luck the new IC's might arrive tomorrow.

 

Did you put fatter capacitors in it?

 

My Ariston amp is now fixed, yay

The new TDA1541A IC's took a little longer than I expected to arrive and I was working in abroad in Spain until yesterday so I only got around to fitting them today.

Here is a pic of the innards.



The Ariston AX 910 shares the same power supply and power amp board as the Cambridge Audio A1 mk3 V2 but the pre-amp section on the Ariston is more basic.

The pre-amp on the Cambridge has two op-amps per channel (NE5532) whereas the Ariston uses a single NE5532P. Also on the Cambridge the pre-amp section benefits from regulated power supply comprising of 4x 7815 voltage regs which provide power to the op-amps. The pre-amp section on the Ariston does not have any voltage regulators and it took me a while to figure out that it uses zener diodes to knock the 30ish volts DC feed from the power supply down to a sensible voltage for the op-amps. Unfortunately I have been unable to find a copy of the schematic for the pre-amp section of the Ariston.

I hatched a plan to upgrade the Ariston by adding a dedicated regulated power supply for the pre-amp board. I used an LM317/LM337 bsed module which I purchased on ebay direct from China in kit form. I'm a big fan these modules, they are ludicrously cheap, very compact, easily upgradeable and generally very useful. At the moment I have set the output voltages to +15v and -15v.

I also swapped all of the electrolytic capacitors for good quality items. I used a mix of Panasonic FC, Panasonic NHG and Rubycon ZLH. I stuck with the same capacitance values with the exception of the eight large filter caps which I upped from 2200uF to 3300uF. I also changed all of small value caps to film types, either polyester or polypropylene.

The op-amps are still the original equipment NE5532P but I have put them in sockets to allow me to change them easily in the future if I want to.

I also hardwired the connections from the input selection PCB to the pre-amp section with silver plated copper wire and swapped the wires which carry the signal output from the amp chips to the binding posts to decent gauge silver plated copper wire.

I've been listening to it this evening and it's really not bad. OK it's not exactly high-end but it has a very energetic character and surprisingly good transparency at times.

All together it's been a fun little project and even with the replacement TDA chips it has only cost me around £40 (including the purchase price of the amp).

Thanks Rob and Simon for the help.

 

It looks good and those amp chips are capable of very clean performance, although not great power. Just visually I'd want to put bigger caps in there, but I guess if it is running on a low'ish voltage then 3300uF is enough. I use 10,000uF in most of my amps.

Very little wrong with 5532s even compared with todays best op-amps.

What is the missing card for? I can see pins but nothing connecting! Reminds me of a CD player I used to have where I could see lights on the LCD for options of the higher-end model. I was convinced it must have the buttons hidden somewhere!

 

The missing card is for an optional phono stage.

The power supply/power amp circuit board has through holes to accept either 2x large capacitors with 10mm lead spacing such as snap-in caps used in my Linn Klout or 8x smaller capacitors.

Originally it had 8x 2,200uF fitted giving a total capacitance of 17,600uF. I could have fitted two large 10,000uF capacitors (20,000 uF) but I chose to stick with the 8x 3,300uF Panasonic NHG caps which I had already fitted which gives me a total of 26,000uF. I can always swap them again later.

I think the Ariston AX-910 had a rated power of 25w so it isn't exactly a powerhouse. A few years ago I had a Linn majik-i integrated which used the TDA1414A chips and that was a very sweet sounding amp though.

 

Ahh that's easily enough. I didn't really look at the fact there were 4 per side they just looked small.

When people talk about total capacitance isn't it usually per side of the PSU, or is it both sides together?

 

It has four 3300uF caps each side of the PSU (13, 200uF total), or eight 3300uF totalling (26,400uF) if you count both sides of the PSU.

The Panasonic NHG 3300uF were the best I could find on Farnell with a voltage rating of 35v (or higher) and with the right diameter to fit on the board easily.

My next job is to do a refresh/service on my Dad's old Pioneer receiver from the 1970's which my sister has started using. All those glowy dials and brushed aluminium should be right up Rob's street.

 

mmmmm glowy dials.......