passive preamp for AKSA (kind of)

rwalton · « **Reply #20 on:** 14 Mar 2003, 04:55 pm »

Following up on the TVC concept, I recently purchased a Sowter-based TVC kit from Horace at iagaudio.com. I have not yet started to assemble it, but it comes with a very nice enclosure and all the parts needed to assemble a complete TVC. Horace modifies the spring on the Shallco 27 position volume switch to reduce turning torque. It works very nicely. This is also the least expensive option I could find for a TVC kit. Only $US 450.00 including a very nice chassis. Check out the pics at iagaudio.com. Horace shows the chassis pics, but he doesn't have kit pricing on the web page, just the cost of a built unit ($600.00 I believe).

I will be connecting it to both my 55W AKSA Nirvana and 100W AKSA amps (but not at the same time). I plan to comment on this combination when I get a chance to do some serious listening. I'm sure it will be quite a departure from my current pre which is an Adcom GFP-555.

BTW, I have no affiliation with iagaudio whatsoever.

Rick

bsy · « **Reply #21 on:** 16 Mar 2003, 08:23 am »

I am interest in Sowter-based TVC kit at iagaudio.com too. It looks great.
Any schematic for the wiring ? Is it US450 for one channel or pce.
Thanks

AKSA · « **Reply #22 on:** 16 Mar 2003, 08:42 am »

I have to say that a transformer based solution is one I'd consider very carefully, as I know that transformers have problems of their own. Most transformer manufacturers will tell you that the best transformer is no transformer!!

I'd also say that a good transformer is incredibly expensive; gain oriented laminations, careful partititioned wiring, etc. To do it properly is very expensive, and I for one would shy away from spending $US450 on a transformer volume control. This sort of money spent elsewhere, with maybe $180 spent on a good DACT or Roancap is probably more cost effective.

I don't want to poor cold water on the TVC idea, many say it's very good, but I do think it can be done just as well with a resistive divider using Vishay high tolerance resistors which sound very nice.

Cheers,

Hugh

Larry · « **Reply #23 on:** 16 Mar 2003, 11:38 pm »

Quote from: AKSA

I have to say that a transformer based solution is one I'd consider very carefully, as I know that transformers have problems of their own. Most transformer manufacturers will tell you that the best transformer is no transformer!!

I'd also say that a good transformer is incredibly expensive; gain oriented laminations, careful partititioned wiring, etc. To do it properly is very expensive, and I for one would shy away from spending $US450 on a transformer volume control. This sort of money spent elsewhere, with maybe $180 spent on a good DACT or Roancap is probably more cost effective.

I don't want to poor cold water on the TVC idea, many say it's very good, but I do think it can be done just as well with a resistive divider using Vishay high tolerance resistors which sound very nice.

Good engineering is to find the weakest link in the chain and put the limited resource to strengthen it first.

I used a Goldpoint attenuator with Dale resistors as a passive preamp for a while. A step attenuator has 3 types of designs. I prefer the shunt to the series type. Note that the signal does not need to travel through the questionable contact point and its arm to the output in a shunt attenuator. (The purest purist approach.

) The ladder type is too expensive to me and it has two suspicious wafers too. Now I am using GK-1 with the Goldpoint shunt attenuator in the place of the pot in GK-1.

I am looking for the new weak link in my whole system chain now and I don't know when I would come to point at the attenuator and kick it out, maybe never.

I fully agree with Hugh on this matter. If I have US$450 to spend, I would not spend it on a single part in this case.

Johnny · « **Reply #24 on:** 19 Mar 2003, 03:16 pm »

Hey, has anyone had a chance to audition the new Welborne Ultrapath, a battery powered, tube buffered "passive" preamp?
(As passive as Aunts Cory's I guess

)
What about the xfmr output- good or bad?
Cheers,
Jy

http://www.welbornelabs.com/ultra.htm

jesserparker · « **Reply #25 on:** 19 Mar 2003, 10:10 pm »

the more i look into this, the more i realize that it is the very same idea behind Hugh's TLP. the tube in that merely buffers the output without adding any gain to it. for those out there who want that "tube magic" i'd say just go for one of Hugh's kits. however, i'm thinking that the big difference in the listening experience so many people seem to agree on between the TLP with the AKSA and with just a passive gain control has more to do with the buffering of the output than the inclusion of the tube.

nothing against either tubes or the TLP, but my system is all solid state, and if perhaps i had built a tube output for my CDP, or used a tube Phono pre or something, i'd be interested in the TLP idea, but as it is, i'd rather stick to one technology on my system, and going all out DIY on an Aunt Corey like pre is going to allow me to customize it all i want.

anyhow, i just wanted to point out to everyone that i just realized that this idea of the buffered pre like "aunt coreys" is the same basic technology everyone seems to be raving about in the TLP. go either route, by the research i've done into it i would say that either way is just fine, and a no gain buffered pre is THE way to go in general. so cheers Hugh, what i looked around and researched for quite some time now to find the ultimate idea for a preamp just turned out to be exactly what you are doing all along. also, the AKSA looks perfectly suited for such an approach.

jesse

DSK · « **Reply #26 on:** 19 Mar 2003, 11:24 pm »

Hi Larry,

You mentioned that you are using the Goldpoint shunt type stepped attenuator. Is this the 20k version, and did you compare it to the Vishay Sfernice cermet pot supplied with the GK-1 by Hugh ? If so, how did the sound differ and was it subtle or worth the extra $$$ ?

Cheers,
Darren.

AKSA · « **Reply #27 on:** 20 Mar 2003, 01:33 am »

Thank you Jesse,

Your observations are right on the money; the concept of a unity gain triode buffer came to me many years ago, and in fact I designed the AKSA with a gain of 32dB and a 47K Zin so that this option would always be viable using a tube.

I'd always had a bee in my bonnet about cathode followers; almost everyone I discussed them with canned them mercilessly, and yet I knew that over the years they had been widely used in some of the very best audio equipment of the 50s and 60s, including the Marantz 7. About five years ago I worked extensively on a friend's 211 SET amplifiers, and we decided to do some work on the cathode follower driver for the output tube. It was very successful, improving the amp sound hugely, and I learned a lot from this exercise (B+ was 900V!!). After a lot of reading I decided to give it a bash for an auto application. I had problems with earth isolation and high voltage on the car application, so set it aside (these problems are now solved however) and then went ahead with the TLP version 1, which operated off low voltage. In the finish I relented and went for higher voltage, and the result was evolution into the TLP2, and ultimately, the GK-1, which is different again.

I do subscribe to Nelson Pass' view that 'If you want a circuit to sound like a tube, you use a tube', but I like to think that elements of the tube sound can be designed into SS as well. Neither do I have the slightest compunction about mixing the two. One must be catholic, even if one is a protestant.......

Cheers,

Hugh

jesserparker · « **Reply #28 on:** 20 Mar 2003, 02:54 am »

Hugh,
thanks for your interest in this post, and also i must say for your very active involvement in this forum at large, its nice to see a manufacture who is so interested in general discusion in these arenas without merely trying to hock his own goods, and also simply so interested in the questions and applications about the systems involving their technology.

i also agree with you that tubes CAN be quite well integrated into a SS system, and have witnessed this myself. i guess its just my own indescribable bias in the "holy war" between tubes and transistors here. i think something about microphonics has just always bothered me.

i like tubes, and in many ways they do trigger that certain emotional reaction in me (a lot of which i think is visual), but for some reason, at this point i want to do it all SS before i experiment with the glass.

again, its always a pleasure to see you here.

jesse

Larry · « **Reply #29 on:** 20 Mar 2003, 10:05 am »

Quote from: DSK

You mentioned that you are using the Goldpoint shunt type stepped attenuator. Is this the 20k version, and did you compare it to the Vishay Sfernice cermet pot supplied with the GK-1 by Hugh ? If so, how did the sound differ and was it subtle or worth the extra $$$ ?

I happen to have a 100k shunt attenuator so I put it in. For a 100kR shunt type attenuator, the resistance starts from 50kR so that it's similar to a 50kR pot or 50kR series type attenuator. For most of the values, resistence increases just a little more than 50kR, less than 52kR at 12 o'clock. I have no problem with this value so that I kept it, otherwise I can always change one resistor, called Rin, in order to have 50kR (similar to 25kR pot) or whatever, though doing this way, it will slightly alter the decibels per step but it's not a worry for me. Being able to replace resistor, it is one of the advantage to use Goldpoint since DACT has begun releasing its attenuators readily made with surface mount resistors. Both of DACT and Goldpoint use the same switches made by Elma (correct for what I saw.)

Pots and attenuators are very different devices in construction and materials. The discrete resistors in attenuators generally have less noises than the carbon or cermet film used in pots. In addition, attenuators have a better channel balance, since the attenuators have 2% tolerance (with 1% resistors) while pots have only 5-20% tolerance.

Therefore, replacing pots in signal path with an attenuator can always have some effects. How much? It will depend on the resolution of the overall system. My feeling is that more sensitive speakers will have better chances to exhibit any subtle differences. (I use 95dB speakers.) If we consider the effect of any component upgrades as subtle, it is subtly obvious after I replaced the supplied pot with the attenuator in GK-1 in my system. I needn't try hard to tell the difference. To describe it, I would say it has better clarity and a darker background, which sounds like the last thin veil has been removed.

I think different people have different formula to assess their investments. The pot is an obvious weak link in the GK-1 signal path so that, to tweak it, the pot is the first on my list.

DSK · « **Reply #30 on:** 21 Mar 2003, 12:20 am »

Quote from: Larry

I think different people have different formula to assess their investments. The pot is an obvious weak link in the GK-1 signal path so that, to tweak it, the pot is the first on my list.

Hi Larry,
Yep, I understand the different design concepts and theoretical benefits (I strongly considered a Goldpoint ladder while building my GK-1), but had no experience with comparing stepped type attenuators and, from what I had read, wasn't totally convinced the theoretical benefits translated to significant audible benefits over the better pots out there (in which I would include the Sfernice pot supplied by Hugh).

Everyone seems to have a different opinion, some have even preferred the sound of exotic "series" type attenuators from TKD to "ladder" types. Still others heard no benefit from ladders over cheaper series types like the DACT. At the end of the day I began to surmise that the type of resistor used (eg. Dale, Roderstein, Holco, Caddock, Tantalum ....) was having a bigger bearing on the sound than the attenuator "type" (series, shunt, ladder). However, because I couldn't readily compare the various types in a single component, I decided to just go with the Sfernice pot for the time being and address this issue later. The GK-1 (with Sfernice pot) was actually such a surprisingly large leap over my previous US$3500 tube pre-amp, that the pot issue has drifted to the back of my mind. However, reading your post, I can feel that itch coming back

I totally agree that tweaks/upgrades are more audible in higher resolution systems. Though my speakers are only 89db sensitivity, they are ribbon hybrids with excellent resolution and inner detail, and have readily shown the impact of tweaks.

The only reason that I was considering ladders over shunts, was that I felt the shunts held more potential traps (due to variable impedance related issues) for someone like me with limited experience in this area, despite the possible advantage of the elimination of one switch contact in the signal path. Even if I got everything right, I worried that it might not remain so if I changed equipment down the track.

Thanks for your reply, please excuse me while I go and scratch an itch

....Darren.

MarinRider · « **Reply #31 on:** 21 Mar 2003, 09:40 am »

I used the Sfernice dual concentric pot some years ago (in fact I used to sell them), they sound clean, clear and tight but just a bit thin and clinical. Mechanically they are poor and tend to fall apart after a couple of years - or the low volume channel balance goes askew (I refer here to the dual concentrics - I don't know which ones Hugh uses and Sfernice's construction may be better in this millenium). However, coupled with a warm amplier stage it will probably sound great.

I now use either the Panasonic plastic type - great sound and a bit pricey, but I have heard they are going obsolete (can any one else confirm this?) or an Alps Blue. The Alps Blue sounds fine and in my opinion the buffer and it's PSU that you use after the pot has a much larger affect on the sound quality than the pot itself. I have tried a fixed attenuator (L-pad with resistors) to see if it's worth putting up with the inconvenience of a stepped attenuator and IMO the difference in sound in marginal.

I have tried all these pots in passive mode with lots of different cables and much prefer the Aunt Corey approach i.e. buffered.

Here's my approach:
Use a passive pre to judge an active pre. A good active stage should have all the benefits of the passive (detail, clarity, good tonality) with better dynamics, rhythm and control of the frequency extremes.
The power supply for the buffer has the most effect on the sound, then the amplifier type, then the pot.

I agree with Hugh on the transformer. My recommendation for anyone wanting to try one is to listen at great length first. If you can't, buy a 1:1 transformer (they are a lot cheaper) and put it in the signal path and have listen.

All the above IMVHO.

Dave

jesserparker · « **Reply #32 on:** 21 Mar 2003, 09:47 am »

dave,

since you seem to have experience with this type of setup, i wondered if you had a look at the schematics for the Aunt Corey pre, and your thoughts on the design. though the schematics are only shown for the initial design stage, the mods are described in detail in the Mo' better mods section. of most interest to me there are the upgrades made to the PSU, making it nearer to pure dual mono.

if you have any thoughts, or considerations for construction/improvement on this design i would be very interested to hear them, thanks

jesse

also, i was planning on using the alps blue series for the pot, so i'm glad to hear you also like the sonics of it.

MarinRider · « **Reply #33 on:** 21 Mar 2003, 10:54 am »

Jesse,

The Alps Blue is like a BMW 5 series. DACTs and Goldpoints mgiht be like Porsches and Astons. (Transformers might be like the new Rolls

The Aunt Corey thing looks good - I have never tried a BUF-03 but would expect it to sound pretty good in this application. The AD744/AD811 combo (Audio Amateur, Pooge 5 by Walt Jung if I remember correctly) sounds pretty good. The GK-1 gets rave reviews so maybe a triode follower is the best buffer .

It is worth checking out DVV's website for PSUs and op-amp suggestions, he knows his beans (and chips):

http://www.zero-distortion.com/glavna.htm

I prefer very fast, low impedance shunt regulators located very close to the op-amps, but not many others use them so maybe I'm missguided?

Cheers,

Dave

mb · « **Reply #34 on:** 21 Mar 2003, 11:27 pm »

Quote from: jesserparker

dave,

since you seem to have experience with this type of setup, i wondered if you had a look at the schematics for the Aunt Corey pre, and your thoughts on the design. though the schematics are only shown for the initial design stage, the mods are described in detail in the Mo' better mods section. of most interest to me there are the upgrades made to the PSU, making it nearer to pure dual mono.

if you have any thoughts, or considerations for construction/improvement on this design i would be very interested to hear them, thanks

jesse

also, i was planning on using the alps blue series for the pot, so i'm glad to hear you also like the sonics of it.

Hi Jesse,

In my experience, battery power is the way to go. It will make things simple, but at the same time you get ultra-quiet supply that's quite optimally matched to the consumption level (<20mA typically). There is no formula for good battery-powered sound that I've seen, so try various types of batteries (sealed lead acid makes sense in size / cost) + caps (size / brand). You will be able to hear the effect of power / bypass caps quite distinctly, and need to give LOTS of time for caps to run in, perhaps because the current load is so low. My battery powered attenuator improved in sound over a period of 2 months... each time you swap caps, it starts from almost zero again.

DSK · « **Reply #35 on:** 24 Mar 2003, 12:26 am »

FWIW, Hugh had the Alps Blue in his GK-1 while he waited for the Sfernice cermet pot to come in. We both agreed that the Sfernice was much better.

jesserparker · « **Reply #36 on:** 25 Mar 2003, 12:30 am »

where can you find a sfernice?

AKSA · « **Reply #37 on:** 25 Mar 2003, 08:46 am »

Hi Jesse,

I sell 'em, and they cost $AUD53 each, which is $US31.50.

I include them stock in all Option 2 GK-1s.

They are dual gang, with single shaft, 22K, log profile.

Cheers,

Hugh

jesserparker · « **Reply #38 on:** 25 Mar 2003, 09:42 am »

thanks hugh,
my only problem with that is that i had wanted a higher input impedance. one of the advantages of this buffered pre is that i can have an extremely high input impedance, and an extremely low output impedance. so i kind of feel that using a 22K or so pot is going to negate this advantage. however. . .
you do a)make buffered pre's, and b)use this pot on them. so i figure there must be some rhyme to your reason and was wondering why you would/do use a pot of that value when you could easily use (and i would think want to take advantage of the fact that you could) a 100 or 200K even pot. thanks yet again hugh,
cheers,
jesse

PSP · « **Reply #39 on:** 25 Mar 2003, 03:16 pm »

Hi Jesse,
The Foreplay preamp has 100K input impedance... it uses a 10K pot (or stepped attenuator for the Sweet Whispers option) in series with a 90K metal film resistor.

Just a thought...

Peter