Page 1 of 3


PostPosted: Sun Feb 11, 2007 2:14 pm
by analog 2.0 studios
We are considering recapping our TAC Matchless.
I've noticed a few threads along similar lines.
Does anyone have any suggestions or feadback?

PostPosted: Sun Feb 11, 2007 2:55 pm
by emtee
Do it :D


PostPosted: Tue Feb 13, 2007 1:24 pm
by nipper
analog 2.0 studios wrote:We are considering recapping our TAC Matchless.
I've noticed a few threads along similar lines.
Does anyone have any suggestions or feadback?

Recap Recap Re...cap
Very important the PSU because is the starting point for noise...
All electrolytics for better frequency response (expecially high and low end) and fatter sound...

PostPosted: Tue Feb 13, 2007 3:08 pm
by analog 2.0 studios
We will be beefing up (replacing) the power supply at the same time we recap. We've only had one minor meltdown, but I feel its due.

PostPosted: Wed Feb 14, 2007 1:42 am
by Audio Maintenance Limited
Obviously we stock all the capacitors for the Matchless.... the type we supply are those used on the 9098i and BC3 consoles which always get very good results.

Let me know if you need a list of the capacitors....


PostPosted: Wed Feb 14, 2007 4:59 am
by emtee
I'd thoroughly recommend ordering the caps from Audiomaintenance - Colin obviously knows these consoles well & can supply the right number of caps for the whole board which saves you going through the modules/schematics & working it out for yourself! (there's one 100uF hiding away between two pots on the MCDS1000 module that's very easy to miss...).

I'm about a third of the way through re-capping my 36 channel matchless and I'm very pleased with the results. The caps supplied by Audio maintenance were very good value for money IMO and well worth the investment.

Make sure you've got a good de-soldering tool as it's quite easy to lift tracks on the boards when re-inserting the replacement caps if the holes aren't clear of solder.

PostPosted: Thu Feb 15, 2007 1:10 am
by brianroth
Since I'll be the "lucky guy" <g> who is doing the recap, I would be interested in seeing the capacitor list as well as obtaining a quote for a complete capacitor kit.

I am no stranger to recapping (or "reswitching" as I did several years ago with a TAC Bullet!) and have a powered vacuum desoldering station and the experience to use it. Hence, my typical approach is to pull the various modules and examine them along with the schematics to create a parts list, then scour the Digikey/Mouser/etc. catalogs for suitable replacements. I always seek "low impedance" caps rated at 105 degrees C. In the case of coupling caps, I almost always increase the value of the caps to "the next highest standard value" to improve the frequency response (and hence phase shift) at the low end, but stay with original values for the rail bypass caps to avoid an even higher inrush current as the desk is powered up.

Colin, interesting that you should mention the 9098i desk since there is a large one here in my city. Recently we discovered it had MORE than a few 85C caps stuffed onto the circuit boards! That was a bit of a surprise.

Back to the Matchless in question. The OEM power supply had been obviously serviced by a previous owner, and I recently repaired a "meltdown" around the front panel fuse holders.

I am suggesting the replacement of the entire PSU with either modules from Acopian if budget permits, or International Power (as found in the 9098i PSU chassis units)/Power-One/etc open frame linear modules that would likely be mounted into a new metal enclosure.

As I recall (but need to verify), the bipolar rails for the audio were fused at 10 Amps, so hence I would use the International 15V/15A modules with the same mods as found on the 9098i to deal with the inrush current "shutdown" as well as the ability to adjust the PSU modules to a higher rail voltage.

Regardless of the final selection of Acopian "Gold Box" or open frame linear modules, I also intend to beef-up the interconnect "snake" between the PSU and desk with multiple runs of heavy gauge interconnect wiring, probably at least an equivalent of 12 GA for each rail.

Within the next couple of days I will be re-examining many aspects of the desk as I prepare my list of materials and labor for my client. Any suggestions are appreciated!

Best regards,


PostPosted: Thu Feb 15, 2007 11:43 am
by Matt Syson
Hi Brian
The fuseholders are / were an issue because the AC current is significatly higher than the DC so larger fuses are necessary. They also have to cope with the switch on surge current. They suffer because they use a front panel type which traps the heat where really an open type would be more suitable for this application.
A properly maintained original supply is fine for the desk but if fitting supplies with much greater current capability then the Imax should be REDUCED so you don't trash the desk when you get a fault.
There is not much point in going too mad with thickening the DC cables as a bit of resistance here also helps with surge currents. Contrary to some myths, a mixing desk is NOT a power amplifier driving a loudspeaker so stupidly big current capability of the supply and power cables makes NO difference to the audio. The current used for the individual amplifier stages comes from the local bypassing caps and the design of the PCB in preventing the stage from amplifying 'interference' signals.
A power supply typically has an output impedance of 5 - 20 milliohms. The interconnecting cable probably 0.1 or 0.2 ohms, the input diodes in the desk a fraction of an ohm (non linear) the isolation resistor in the module 4.7 - 22 ohms. Doubling the impedance of the supply therefore makes virtually no difference.
A bigger HEATSINK so that the supply runs cooler makes sense, overall current capability (beyond the minimum necessary) is largely irellevent.
Matt S

PostPosted: Thu Feb 15, 2007 3:16 pm
by brianroth
Interesting post, Matt! I was working from memory with 10A rails in mind, and the closest open frame linear module with at least a 10A rating was a 15A unit, the same as used in the 9098i desk.

As for the wire sizing, I know that 14 gauge is rated for a maximum of 15A when used for AC mains applications in the USA (and, any mains branch circuit is not supposed to be loaded to more than 80% of the breaker's rating). The length between the power supply unit and Matchless desk is currently in flux, so my idea of ensuring a hefty wire size.

Speaking of hefty wire sizes, the cables linking the 15A supplies to the 9098i desk are in size 0 or larger. They have the appearance of battery cables as found in a truck.

I will definitely review the Matchless schematics to see what (if anything) lies between the output of the supplies and the rail pins on the opamps.


PostPosted: Thu Feb 15, 2007 4:55 pm
by Matt Syson
Don't count the 16mm squared wires on many of the AMEK desks these were fitted mostly to 'standardise' and gain a decent price for bulk purchase. 10mm square or even 6 would have been fine. The object is not to lose more than about half a volt on the run. Many of the supplies on Mozarts, Hendrixes and others used 2.5mm INSIDE the box, about 1.3 metres of it. Having 8 metres (the standard shipment) of 16mm was a nonsense. I had the internal wiring uprated to 2 pieces of 4mm for ease of wiring. This is still significantly more than that outside the box.
The Matchless uses diodes as isolation between modules, non linear and impedance related to the current.
Aim for about 0.3 volts drop per wire and you will be fine.
The current rating given for power cables is interpreted from the expected drop and legislation about maximum drop rather than the point at which it melts.
Desk manufacturers specify cable sizes and make strong arguments against changing them because the cost of 3 days fully testing a particular desk with different cables would remove all the profit.
Matt S