30mA would equate to approx 13mV across the pair of 0R22 resistors wouldn't it? Sounds about right, but this is a little higher than is recommended for the NAC 90.3 amplifier (10mV I believe), albeit with different output transistors.
Other posts on this site suggest that 20mA is more appropriate though:
http://www.diyaudio.com/forums/solid-state/172509-upper-limits-biasing-2sc2922-2sa1216.html
Last edited:
Naim designs of this period and earlier are all quasi-complementary. Consequently, the bias can never be set at an optimum level for both EF and CFP devices, which might otherwise be set at about 14 mA for the CFP and about 35 to even 100 mA for the EF device. I always end up with 30 mA bias and other, wiser heads claim that the bias current is non-critical from 10-35 mA - within the range of models considered here: Modifying Naim Audio power amplifiers
Predictably, some folk will want to fuss over bias current, perhaps because they either feel it should be specified somewhere or they prefer some particular level for their idea of good sound quality. I've listened over long periods to a few original models and I can't always tell if there is a real or imaginary difference between 10 or 30mA bias but that issue is more about bias stability, which at low settings like 10 mA, will vary more with ambient temperature than that of the output stage - not a good situation.
Whatever the bias levels recommended for EF output stages using complementary pairs, 2SC2922 pairs have been used in the NAP140 and several other NAP designs for many years with no different bias recommendations to that with other transistors like BDY56. BDY58 and NA007, which have also used at one time. The 350W NA007 device is more than likely the NPN type from the unique 350W complementary pair in the Semelab transistor range.
Predictably, some folk will want to fuss over bias current, perhaps because they either feel it should be specified somewhere or they prefer some particular level for their idea of good sound quality. I've listened over long periods to a few original models and I can't always tell if there is a real or imaginary difference between 10 or 30mA bias but that issue is more about bias stability, which at low settings like 10 mA, will vary more with ambient temperature than that of the output stage - not a good situation.
Whatever the bias levels recommended for EF output stages using complementary pairs, 2SC2922 pairs have been used in the NAP140 and several other NAP designs for many years with no different bias recommendations to that with other transistors like BDY56. BDY58 and NA007, which have also used at one time. The 350W NA007 device is more than likely the NPN type from the unique 350W complementary pair in the Semelab transistor range.
Last edited:
Thanks, the case arrived today and I'm impressed as you alluded to.
Annoyingly they've sent the version with a vented top even though I paid extra for the lid to be solid. The aluminium bars have also arrived so construction of the heat transfer assembly is underway.
Getting close now, just waiting on the transformer...
Hi, the recent discussion has been around the NAP 200 clone. We seem to have have taken over this thread with NAP 200 discussion.
A couple of us have nearly completed the NAP 200 and all is looking good so far. I don't think it will be too long before between us we have one up and running.
Chris
If you are particularly interested in Naim products and their unique sound quality that belongs to a past era, then the NAP 200 kit currently being discussed, is absolutely the best quality and most original kit to date, IMHO. All NAP amps from the period use a very similar design and meticulous layout so they and are comparable in most respects. Though much more expensive, this NAP200 kit looks very promising though no one here has yet completed it.
If you mean the remaining two NAP 140 kits by Hifidiy and LJM, still widely sold on Ebay, then you have a choice - One is close to the original Naim power amplifier PCBs but comes with many dissimilar components. The other is a cut down, simplified version that is even less like the original. You could ask, why bother trying to make it perform like the original, when you need to replace most of the semis? It could be used as a board to make clones of spin-offs like Avondale's NCC200 if you wish, but really, that deserves another thread.
In their favour, these NAP140 kits are cheap and fun if you bear in mind the correct power supply is important and is only +/- 34V for NAP 140, but you can have great sound using the original parts where necessary. This has been discussed over and over in the thread. Go back a couple of years and start where the discussion is still on-topic. Take a look at the pics early in thread - what do you think for the money?
Last edited:
All the magic is in the TR2 collector resistor. You can add this to pretty much any power amp and get a taste of the best aspect of the Naim sound. The 0.22R output resistor that Naim use is not to my taste, but it's easy enough to replace with a coil/resistor.
1/2 the sound comes from the power supply so I suspect most of the clones out there will sound different (and different to a Naim original).
Whether or not it's worth building is up to you.
Getting closer... just awaiting the transformer, voltage regulator heatsink and a few bolts (and some spare time).
The heat transfer assembly is now ready, just need to drill some holes for the PCB standoffs and then for the connectors that will mount onto the back panel.
Chris
Attachments
Last edited:
Not really, try swapping ZTX653/753 for the common kit substitute parts like 2SD697/B647 and listen again. merlin2069er is talking about about kits, no?
Compensation and the VAS also play a major role in stability and determining whether the SQ is poor, ho-hum or really good. Certainly, many components are unimportant for any sort of end product you may decide to clone instead but if it's Naim sound you want, I much prefer the original component lineups with the possible exceptions of the output and input pairs, input current source and Vbe multiplier transistors.
How that applies to other variants like the NC200, is a different consideration but I've built, repaired and corrected enough poor supplier substitutions and newbie mistakes on the NAP clone kits to be clear on this.
Turns out it was my fault. The case apparently comes with two vented panels as standard, I'd specified that only one should be solid. Never mind, I'll just have to seal off the vents on the top panel somehow.
I had an early NAP140 kit a few years ago and I think that had something other than ZTXs in the VAS and corresponding current source. it sounded poor.
Certainly my NCC200s were built with all the recommended bits (including the becoming-fabled BUV20s). The reason I assumed that the ZTXs were not part of the magic is because my 002CCN uses MJE340 for the VAS and 2SA1360 for the current source and it sounds rather better than the NCC (using the same PS).
Anyway, it looks like you have more experience with the NAP circuit than I do, so maybe my assumptions are a step too far. All fun and games
Someone said brass might be poor for a heat transfer device. Yes and no. The surprising thing is the transfer pads really are poor. Some amps have all the NPN on one heatsink and PNP on another. The heatsinks electrically isolated from the case. This allows just heat transfer paste to be used on the power transistors. A&R A60 had very high grade anodising on the heatsink. It managed to use the same trick without the complex isolation. That's brave and it worked.
The transformer has just arrived and looks pretty impressive. To look at it's a lot bigger than expected, measures around 160mm in diameter and 70mm tall, but looks as though it will fit into the case (ModuShop, part no 1NSLA02350N). I've attached an image, but the image doesn't really portray the size of the thing in real life.
Requested spec was as follows:
Power rating: 450VA
Primary: 0-230V @ 50Hz
Electrostatic screen
Secondaries: 2 x 28-0-28V @ 3.57A rms + 26-0-26V @ 0.96A rms
GOSS band
Regulation: approx 5%
Estimated dimensions: max 75mm high
Mounting: insert in a potted centre
Leadouts: approx 400mm long flexible (stranded)
The transformer was built and delivered within a week (Tiger Toroids Ltd in the UK) and was quoted at £75 + £13 carriage + VAT@20%. They suggested using a 600VA transformer core, a lower-rated core may well be cheaper but I didn't pursue that.
Over the next couple of nights I'm expecting to be able to power the amp up for a first listen. Fingers crossed.
Chris
Attachments
Andrew. What fusing do you think a 450VA will need? I know it is something you have given much thought to in similar discussions. I suspect 4AT or larger. It is a difficult question as 4AT relates to about 1000 VA and yet might be too small. If avoiding a slow start circuit I guess that's about right at 4AT. I am very reluctant to open this can of worms. Feel I should. 4AT@230V will protect against the bad risks.
One thing that might be possible is use 2.5AT and accept like Naim you never switch off. Not my favourite.
Doubtless Tiger will give an idea. If the transformer has a heat fuse one can be slightly less concerned. Sadly if a one shot thermal fuse it would be a shame to loose the transformer if the Naim module went wrong. 2 Fuses in the DC supply seems wise. If these are bypassed with 1 uF ( across the fuse ) that is almost like no fuses fitted.
One thing that might be possible is use 2.5AT and accept like Naim you never switch off. Not my favourite.
Doubtless Tiger will give an idea. If the transformer has a heat fuse one can be slightly less concerned. Sadly if a one shot thermal fuse it would be a shame to loose the transformer if the Naim module went wrong. 2 Fuses in the DC supply seems wise. If these are bypassed with 1 uF ( across the fuse ) that is almost like no fuses fitted.
I'm starting with a T3.15A in the mains feed, as found in the genuine NAP200. The amp will be running on 230V of course.
Random image from Google image search:
http://tunehifi.com/wp-content/uploads/2015/01/Naim-NAP-200-Power-Amplifier-rear.jpg
In Naim's description for the NAP200 they boast "A new, top-quality 430VA toroidal transformer", T3.15A should be OK.
If the DC rails were individually fused then what would be the effect of just one of the fuses blowing? If there was to be resultant damage to the attached loudspeakers then I'd rather lose the transformer due to the thermal fuse tripping.
Chris
Last edited: