People talk about the unregulated Naim amps having a certain 'something' that the regulated amps seem to lack. Most notable are the 110 and 160 in this regard - both possessing a similar style that is difficult to describe but (when you hear it) equally difficult to do without. Let's call it Mojo.
For too long now my olive 250 has been boxed up doing nothing in favour of a 110, and even though the 250 has better bass control than the little amp, the 110 kills it in the all important midrange. Given the two amps use the same output boards, I figured this really has to be down to some weakness in the 250s regulator boards, and in particular, the interface between the regulators and amp boards. There is just 10uF of capacitance per rail at the O/P of each regulator, so there must be considerable noise on the power rails - something that would likely be audible in terms of loss of low level information.
I decided to try adding some extra capacitance, fitted at the amp boards, and with the emphasis on using only the minimum of extra capacitance to get the required result. A bit of searching revealed this has been tried before and was not entirely positive but having experimented with various values I feel they may simply have used more capacitance than was optimal.
The optimal value seems to depend on the load the amp sees. An easy load like ES14s requires as little as 47uF per rail to hit the 'sweet spot' to my ears. In fact anything up to 100uF is very good with the 14s and is probably a value that would work well with about 90% of speakers out there. The Gale 401s are a heavy load compared to most and the sweet spot for them appears to be 220uF so that's what I've settled on for my 250. The difference this has made is that the 110 is now relegated to workshop duties.
'nuff said
Mr Tibbs
For too long now my olive 250 has been boxed up doing nothing in favour of a 110, and even though the 250 has better bass control than the little amp, the 110 kills it in the all important midrange. Given the two amps use the same output boards, I figured this really has to be down to some weakness in the 250s regulator boards, and in particular, the interface between the regulators and amp boards. There is just 10uF of capacitance per rail at the O/P of each regulator, so there must be considerable noise on the power rails - something that would likely be audible in terms of loss of low level information.
I decided to try adding some extra capacitance, fitted at the amp boards, and with the emphasis on using only the minimum of extra capacitance to get the required result. A bit of searching revealed this has been tried before and was not entirely positive but having experimented with various values I feel they may simply have used more capacitance than was optimal.
The optimal value seems to depend on the load the amp sees. An easy load like ES14s requires as little as 47uF per rail to hit the 'sweet spot' to my ears. In fact anything up to 100uF is very good with the 14s and is probably a value that would work well with about 90% of speakers out there. The Gale 401s are a heavy load compared to most and the sweet spot for them appears to be 220uF so that's what I've settled on for my 250. The difference this has made is that the 110 is now relegated to workshop duties.
'nuff said
Mr Tibbs
