Un…..fortunately…..we have yet another example of a company on the Unofficial Secrets list. This means that their designs are so clever that nobody else should be allowed to repair them.
This could also mean, possibly, that it would be unwise for them to let anybody in there who might know what they’re doing. Because they might just sus out that these designs are not particularly clever, nor are they especially well manufactured, nor do they work all that brilliantly well.
It has to be said that this amp was devastatingly silent in operation.That’s because it didn’t work at all.
These two little black blobs are TDA7294 integrated circuits (chips), and these are the devices that make this little AER amp make sound. Each device is rated at 100 watts rms dissipation, so should, on the face of it, be good for 200 watts all told. Each chip supplies one of the speakers. So far so good.
The property that limits the output of these chips (and also causes them the most problems, often) is how fast they can get rid of the heat that is developed in pushing out power to the load (speakers). These are exactly the same chips as are used in the Marshall AVT series amps. The AVT 150 uses two of these in bridge mode, which is something I should cover as we go on. So back to that in a bit. The difference in design is not great, the major difference being that in the Marshall amps the chips are built on a separate pcb and is fan cooled; which is a big plus in getting rid of heat. This why the Marshall AVT150 has the same two chips rated at 150watts rms and the Domino is rated at 50 watts lower. In the Domino amp the heat sink is alloy and has no forced draft cooling so the heat dissipation characteristic is much lower.
A major reason why this amp decided to self destruct was a result of the heat sink integrity. The stuff that is used to ensure that the heat is carried away very quickly is ’heat transfer compound’. This grease-like stuff you apply to the heat sink/ power amp chip surfaces and it keeps the temperature gradient across the junction of the heat sink and chip surface to a low value. Un…fortunately this stuff can harden with age and can become a heat insulator; so it has exactly the opposite effect as it was supposed to have when first manufactured. If you’re going to get into a big service on one of these things, one of the operations should be to release the heat sink from the chips, clean the surfaces and re-compound the thing.
The fault on this particular little Domino was fairly obvious straight away; a leg had blown off a power amp chip. The fix is to take the TDA7294 out (both of them preferably) and replace them. This is not as simple as it might sound. Desoldering 15 pins (30 if both chips are removed) on a modern printed circuit board is a tricky operation. The reason is that if you get the tracks too hot they’ll peel off. If you don’t do a great job on desoldering, you will pull the tracks/pads off. Also, as the pads are close together there is a possibility of solder bridges. Be very wary of all this stuff. The approach with the least trouble is to take the fixing screw out of the TDA chips and lever it away from the heat sink a little. Then cut all the pins. Resolder the pads before you desolder them. That’s because when robots solder joints they tend to have little tin left in them, which means you are desoldering lead, which means you have to get the joints hotter which means that you might start to lose tracks. Bad idea.
After that you need to make sure you are grounded. With a wrist strap or some such. Static is a killer on these things. The chips themselves are not a lot of money but the job itself is a fiddly one.
Now we get to the hair pulling out stuff. All back together, this amp worked fine on the bench. I ran it up to a decent output on dummy load checked everything was getting rid of heat as it should, and it went out.
It came back. It hadn’t worked at all for Tom. And I put it back on the bench and…..it didn’t work for me either. Bugger. I couldn’t find anything wrong, so I decided to do the whole job again. For nothing, of course. Only engineers are stupid enough to be engineers. I must have said that before somewhere. But it didn’t work. Not a dickey bird.
As this was a produce of the AER Secret Society, I couldn’t find a schematic for it anywhere. (Does the word Samson spring off the page here????) So I downloaded a TDA7294 data sheet. You can find these without much trouble. It at least gave me a bit of a handle, and as we know by now, I really hate the so-called ‘Free Information Society’ mostly because it’s bullshit. So I will go to a lot of trouble to show these folks up for what they are. Anyway, onwards and….whatever…..
The data sheet showed that two of the fifteen pins (9 and 10 if I remember right) were used as a standby and a mute. If the mute pin had a low voltage to it the amp wouldn’t do anything. If the standby had a low on it, it remained in standby. Well I had to draw out a bit of the circuit, but I found that it was just a capacitor that charged up to put a delayed high to the mute pin. In this pic the grey/black cap to the right of the red probe was leaking badly. This is common to both TDA chips and so mutes both amps at switch on. Un….fortunately….if the cap is leaking the mute remains on. That took some figuring out, no thanks to the tech support of AER.
Just as an aside, I mentioned ‘bridge mode’ earlier on. Many power amps have a bridge mode setting and you need to use it with a lot of caution. The arrangement uses two power amps. In a stereo power amp these would each drive a separate load; the amp drives into one side of the speaker, and the other is connected to ground. In bridge mode, one amp drives into, say, the positive speaker pin, but the other instead of connecting to ground, is driven by the other amp. Unless the phase of one amp is reversed, the result of that would be nothing at all. But one amp has its phase reversed (in the preamp section) so one amp pushes when the other pulls. The result of this is that the power of both amps is summed in the speaker and the power out is equivalent to the sum of the power of both amps.
It’s not impossible to run one of the amps into a short in this arrangement. Hence the gentle warning.
Time for tea.