advertisement


Cambridge Audio 340A SE preout conversion and other mods.

Any chance you'd put the blackgate in the wrong way round? What voltage rating are they compared to the original caps? have you bought a multimeter yet? :)
 
James Evans said:
Any chance you'd put the blackgate in the wrong way round? What voltage rating are they compared to the original caps? have you bought a multimeter yet? :)
Click to expand...

To be honest Im not sure. The stock caps wer small rectangular grey things and I dont understand the writing on them. I guessed they were bipolar as there are no markings on the PCB. I studied the circuit to see which pad led to grond and fitted the BG's accordingly (I at leat I think I did, I might have got it wrong). The BG's I fitted were 16v rating (15v supply to this area of the circuit). Ive got the originals back in and everything is fine.

Still dont own a meter, shoot me! I WILL get one soon! HeHe!
 
mikesnowdon said:
To be honest Im not sure. The stock caps wer small rectangular grey things and I dont understand the writing on them. I guessed they were bipolar as there are no markings on the PCB. I studied the circuit to see which pad led to grond and fitted the BG's accordingly (I at leat I think I did, I might have got it wrong). The BG's I fitted were 16v rating (15v supply to this area of the circuit). Ive got the originals back in and everything is fine.

Still dont own a meter, shoot me! I WILL get one soon! HeHe!
Click to expand...

Hi Mike

Those grey caps will have been film caps, probably polyester. A 16v rated cap is pushing it on a 15v supply. The +ve side of the cap is always to +ve voltage, so -ve to grnd on a +ve supply line, +ve to grnd on a negative supply line.

If you change the caps again I'd go for higher voltage rating, 25v at least.

Glad its working fine now.

James
 
Ahhh! That explains why only one cap turned into a baloon! I will order a pair of BG 25v soon. I assume 220uf will be ok in this position. (Feel free to advise if thers a better cap to use in this position)

Im looking into a discrete class-A buffer module at the moment. Its called a Jisbos. Id like to replace the whole 'input buffer' stage with a pair of the modules if possible. Im waiting for the designer to come back and tell me if they're compatible.

Details: http://www.diamondstar.de/jisbos/jisbos_gallery.html

If the above isnt possible then I'll upgrade all resistors and caps around the opamp. (Of course, teddyregs will be installed eventually too).
 
The Jisbos has a fairly high current drain (30mA), so you need to make sure that the +/-15V suppy of the 340A can handle a pair of such hungry beasts.

It looks a perfectly reasonable circuit. Some of the parts (2SK170, 2SJ74) can be hard to get hold of.

My other concerns would be size (fairly big board, is there room for two?), and the very high loop gain, which can give all sorts of issues with RF stability. You would definitely want to have access to a scope and signal generator to make sure it was working right.
 
PigletsDad said:
The Jisbos has a fairly high current drain (30mA), so you need to make sure that the +/-15V suppy of the 340A can handle a pair of such hungry beasts.

It looks a perfectly reasonable circuit. Some of the parts (2SK170, 2SJ74) can be hard to get hold of.

My other concerns would be size (fairly big board, is there room for two?), and the very high loop gain, which can give all sorts of issues with RF stability. You would definitely want to have access to a scope and signal generator to make sure it was working right.
Click to expand...

Parts are no problem, theres a company that sells Jisbos kits. As for current drain, the best soloution would be a dedicated psu, and it would ensure they were performing optimally (of course this would also benefit the poweramps). I reckon thres plenty of space too, I could probably fit a longish sink down the left side and mount the jisbos on it along with the Teddyregs that would feed it. Of course this is all 'pie in the sky' at the moment as I have no idea about scopes or signal generators!
 
The designer 'Ti Kan' has been in touch:

"Yes, JISBOS can be used to replace the NE5532 input buffer. You will
lose 6dB of gain because JISBOS is unity gain, but this may not be an
issue. Be sure to mount the JISBOS boards as close to the original opamp
as possible, in a secure manner, and use the shortest wiring possible.
Also be mindful that JISBOS will draw more current from the power supply.
If U8 and U9 (7815 and 7915 regulators) get too warm you should put heatsinks
on them."

Good news then! I think the cd player has enough gain allready that the above would indeed work. My only concern is the DVD player, as is I have to turn the volume knob about 10 degrees further than with the CD for the same level.

Any thoughts?
 
Ive had lenghty discussions with the designer of the Jisbos and it seems that the Jisbos would be ideal. The issue of it being 'unity gain' wont be a problem as the -6db loss shouldnt make much difference as teh conttrol range on the pot is plenty big enough. The Jisbos runs of a +/- 15v supply so I can use the existing supply in the Amp. Eventually I'd have a pair running on Teddyregs fed from a dedicated psu. sizewise theyre ideal and not too expensive, theyre also easy to build. As far as I can tell they dont need to be scoped or anything like that. Probably a nice idea to fit a small sink down the right side of the amp and mount the Jisbos and Teddys on it. This is all planned for the future as I need to learn a bit more about amps I think. Im looking at scopes on ebay and I hope to have a meter soon.

I never had big plans to mod the amp, but the mods to the CD player have been so good I feel the need to improve the little 340. Obviously if money wasnt an issue I'd just get a better amp, but modding is cheaper and more rewarding anyway. I dont know when I'll get around to doing the Jisbos, I have a couple of other things on the agenda first. Im going to re-cap the supply with Blackgates and change the decoupling caps (any suggestions?).
 
Naim use a 100uf tant in the feedback loop of the 250 power amp. I prefer a film cap (47uf MMK polyester)
Polypropylene should be even better than polyester and you may be able to get away with a smaller value (polypropylene caps are huge compared to electrolytics of the same value)

One of the more experienced bodgers (maybe Piglets dad or Martin clark) may be able to give you an idea of what value film cap you could get away with.
You could just fit some flying leads to the circuit board to make it easy to experiment (taking the board out every time you try a new component can get rather tedious)
 
Nice one.

I'll do as you suggest and do some rolling to see what sounds best. I'll probably start with the decoupling on signal path. Once Ive found which : cap I prefer I'll try different caps in the feedback. As a first postulate I'll probs have the Nichion Muse as decoupling as I have enough (ordered 10) to replace all of the signal decoupling caps. After that I'll try the Rubys, Tants, and Muse in the feedback.

Im also going to experiment with the opamp supply decoupling and see what differences (if any) that brings.
 
For the signal path (coupling) i like BC128 or Polycarbonate in my Naim gear (Naim use tants). In my Onix i use MMK (polyester).
I would try to keep electrolytics out of the signal path (apart from BC128).
Have a read of www.acoustica.org.uk for a list of coupling, decoupling and feedback caps to try. The list is intended for Naim preamps but will work in any equipment.
 
Well I guess what I have will be an improvement on the cheap lytics allready in the amp. I read the article, thanks for that. Ill do some mix and match with what ive got and see what happens.
 
I've now got a bunch of 2.2uF Mundorf MKP's and pair of 220uF BG STD's. Also got a LM4562 opamp to try (and socket). Im trying to work out which caps to change and what to change them to...

Are 100uF (25v) Nichion Muse any good for local powerail decoupling? If so is it worth upgrading the caps on the input buffer and selector chip?

Obviously all the signal decoupling caps could now be changed to Mundorf MKP's. As I still dont own a meter (groans audible) I will play it safe and chnge all of them.

The BG STD's will replace the 100uF lytics after the +/- 15v (opamp/selector chip) regs.

Could I change the smoothing caps before opamp regs (currently 1000uF on + rail and 470uF on - rail) to 4700uF 25v Panasonic FC?

It looks like the power chips dont have a cap in the feedback, instead there is a pair of resistors. However, on the buffer opamp the feedback caps are ceramic's. Does this mean its worth changing them then? If so what type/value? (I have Tants, Rubycons, MUSE I could try here. All 100uF 16/25v)

Balance pot. How do I get rid?

Thanks in advance.

Mike
 
Well, I'm about to get started. If you see anything obviously wrong in the above post please let me know. To be honest I was hoping for some support/feedback but I'll have to just take a chance and hope im not wasting my time today with these mods.

Rgds,
Mike.
 
Input buffer mods are now done. LM4562 on socket, Muse +/- decoupling caps, Mundorf MKP's for signal decoupling. I had to move a couple of resistprs under the pcb to make room for the Muse. Also repaired a few traces which were damaged on a previous mod. A few more things to do now. Can someone take a look and tell me if they see anything out of place please?

l_fbb1ef7de5a6426082294e8a8a1d8e2d.jpg


l_e8e1b068499245f683312728252ad8ec.jpg


l_63aeb914cdf44448adf30527c35e9090.jpg
 
You must log in or register to reply here.


advertisement


Back
Top