Meridian Unplugged - Powered by MrTechGuy.co.uk Forums Discussion Meridian - Systems Power supply upgrade

For the record, the 5-series equipment used linear supplies, and so did the G01, G02 and G07, but all the other G series and 8 series used SMPS - even the 808.3.

Incidentally, the ECM40UT31 PSU in the Octopart link above is an SMPS where the line regulation, load regulation, ripple & noise are all 1%. Same as every other modern SMPS.

Looking more closely at the secondary supply board in the 861, it looks like that's just a filter. It only has passive components - there are no secondary regulation stages.

I have a lot of faith in Meridian when it comes to developing for the best sound quality (if not ease of installation or customer support). I can't believe they would have compromised such large, heavy and expensive processors such as the G68 & 861 by fitting SMPS, when a linear PSU might have been better. However, the margin of superiority of linear over switching supplies surprised me, and I'm definitely going to try it out.

Simply modifying some equipment, then listening to it afterwards (perhaps a long time afterwards) isn't a reliable way to hear any improvement. If you want to hear an improvement, you probably will. A blind test of two units side-by-side is much more challenging - and convincing. To this end, I've gone mad, and acquired another 861 for experimentation. This one is a V2, rather than a V4, but I thought it might make an instructive comparison. I preferred the V4 to the V2 in previous shoot-outs, so the challenge is to make the V2 sound better - I think that would be very meaningful.



After a few listening sessions I thought I�d capture my thoughts on base-lining the stock 861v2 with the 861v4, before I forget everything. Using the v4 as the reference is important, so I can be confident in judging whether the power supply has any impact on the v2. If they sounded the same before and after, the comparison would be worthless.

And to begin with, they did seem to sound the same. I tried to set them up identically, and even used the same output cards � OA02. These seem to have a significant bearing on the sound, certainly on a superficial level � tonal characteristics etc. When you take the D to A conversion out of the equation, the differences between processors are more subtle.

I think this is why ABX tests are often poor discriminators, and make everything sound the same. If you keep hopping and skipping between equipment, you don�t get to dig into the core of the music, in the way that you can with extended listening. And it was only after a longer time that the improvement of the v4 became clear.

The v2 certainly sounded good, but the presentation was flatter, more forward and two-dimensional. The v4 had more resolution, and initially seemed more analytical, while the v2 was a bit more superficial, but had good energy, drive and rhythm. However, the v4 was at the same time more relaxing and more exciting, and had better micro-detail, so the soundstage was deeper, and better fleshed-out. If playing music is like painting a colour-by-numbers picture, then the v4 was using more colours.

Its getting interesting, but there's more to come...

Nick