Arkless Electronics Transconductance phono stage

Hi-Fi discussion
Post Reply
Arkless Electronics
Posts: 64
Joined: Mon Oct 16, 2017 5:54 pm

Arkless Electronics Transconductance phono stage

Post by Arkless Electronics » Thu Oct 19, 2017 8:19 pm

Hi folks. It seems my attempting to sign up worked in spite of my cocking up and putting .com instead of .net in my email address.... :roll:

Well I was invited by Si of this parish to join up here and continue the saga of the Transconductance phono stage, which of course works on the same fundamental principal as The Paradise.

The present state of play is that the boards (veroboard for prototype)) for the two channels are built, set up and working and it now has its own regulated linear PSU rather than being temporarily powered from my lab PSU.

The discrete buffers were designed and fitted also, as I was getting to in the AOS thread.... They actually do sound noticeably better than the op amp buffers interestingly enough.

It's sounding good and with, I'm sure, more to come from it yet with a bit of further refinement.

Unfortunately the need to earn money has meant it's pretty much on the back burner for now (I may get the odd tweak in :) ), which is precisely what I was wanting to avoid by putting all resources into finishing it, so it could be doing the rounds of those interested in hearing it (and hopefully ordering one) whilst I then dealt with the backlog...
It's a complicated beast, designed from first principles (as you will look long and hard to find examples of "prior art" in transconductance phono stages, other than the Paradise... and this uses completely different circuitry to even that) and it just went way past the time it was possible for me to spend on it (and I really stretched it!)... For now anyway 8-)

YNWaN
Site Admin
Posts: 3023
Joined: Wed Apr 27, 2011 2:02 pm

Re: Arkless Electronics Transconductance phono stage

Post by YNWaN » Thu Oct 19, 2017 8:36 pm

Welcome Jez and I really do hope you are able to bring your design to market.

As you say, the transconductance circuit has very little prior art. However, past experience of such an architecture and the more recent Paradise phonostage suggest it is very much an area worthy of investigation - but requires a genuine engineer prepared and capable of working from first principles (like yourself). For once this is a design that plays to your strengths - it’s also an inherently an expensive thing.
Mark

Arkless Electronics
Posts: 64
Joined: Mon Oct 16, 2017 5:54 pm

Re: Arkless Electronics Transconductance phono stage

Post by Arkless Electronics » Fri Oct 20, 2017 7:11 pm

Thanks Mark.

YNWaN
Site Admin
Posts: 3023
Joined: Wed Apr 27, 2011 2:02 pm

Re: Arkless Electronics Transconductance phono stage

Post by YNWaN » Fri Oct 20, 2017 8:48 pm

Cool but blimey Jez, lets have a bit more content mate, your thoughts regarding the theory and extrapolation of such, photo’s of the thing itself, even your justification for the use of this topology over another (actually, that would be very interesting to myself).
Mark

YNWaN
Site Admin
Posts: 3023
Joined: Wed Apr 27, 2011 2:02 pm

Re: Arkless Electronics Transconductance phono stage

Post by YNWaN » Fri Oct 20, 2017 8:54 pm

Jez, I’m confident that you will be able to design a circuit of real worth - hopefully you, Si, me and our friends will be able to source/design a case that will do it justice at a price that works for you and your potential customers.
Mark

sq225917
Posts: 617
Joined: Tue Jan 22, 2013 9:28 pm

Re: Arkless Electronics Transconductance phono stage

Post by sq225917 » Fri Oct 20, 2017 10:42 pm

jez, get it ship shape and I'll try and get over next weekend. I've just finished my own Paradise PSU so we can give it a good compare and contrast. What mains voltage do you have, i'll trim the psu to suit.
One more Paradise left to complete.

Arkless Electronics
Posts: 64
Joined: Mon Oct 16, 2017 5:54 pm

Re: Arkless Electronics Transconductance phono stage

Post by Arkless Electronics » Sat Oct 21, 2017 3:18 pm

I'm rather reticent about even discussing the basic outlines of this technology, due to it being so "off the beaten track" that virtually anything I say would be useful to potential competitors.... I'll say a little though.

I guess that if it can be said that a transconductance phono stage (TPS) has a big advantage in any way then it is that it uses passive RIAA EQ without the big problem of headroom and overload usually inherent to passive EQ. It completely circumvents the issue.

Another potential plus point is that by its very nature a TPS has no negative feedback. To get the sort of gain, bandwidth and output levels required without incurring very significant distortion takes quite some doing if it is to be accomplished without negative feedback!!
Many believe that negative feedback squashes dynamics, adds "greyness" and "sterility" to the sound and tends to cause very fine detail to sort of disappear beneath an "inaudible noise floor" (i know that "inaudible noise floor" doesn't really make sense but I can't think of a better way of putting it...) and there does seem to be something to this belief in my experience. Just this subject alone could be a huge thread on its own of course!

As I said in the original thread on AOS I independently "invented" TPS's around 20 years ago and have built several different models since then... The hybrid phono stage Mark and Si both tried and liked is a TPS for example and I have in fact used the technique in most of my phono stages other than the Arkless 640P for a long while.
The big difference between all the versions I've built previously and the present one (and the Paradise), and that made it so difficult, is that the Paradise and this new one of mine takes the signal right up to its full output level before EQ'ing it. It is then buffered to allow it to drive a load (it inherently could not drive anything at all without the buffer).
My previous designs have only taken the 0.5mV or so from the MC cart up to around 30mV by the TPS/EQ'ing technique and have then simply amplified that by around 20-30 to give the final line level output. I still believe that ultimately this method has more mileage in it than the technique used in this new one and in the Paradise....
That hybrid one that sounded very good indeed had a fatal flaw which I suspected may come back to bite me and indeed it did... good thing I didn't sell any! I used a pretty bizarre technique to get distortion levels well below that theoretically possible from the second stage valve (which amplified up the already "TPS'ed" and RIAA EQ'ed signal). The problem was that, as anticipated, it was so sensitive to the valves parameters that even 20 hours use was enough to need a return to the bench for adjustments which required lots of test gear... No matter as I soon redesigned the second stage to completely get around this, reduce the somewhat excessive gain and probably sound even better still.... I just haven't got round to doing it in practice as yet!

Disadvantages of TPS's are mainly to do with the difficulty in making one switchable between MM and MC, other than by building one as a MM stage and having a built in head amp which can be switched in for MC use. Personally It doesn't really bother me as IMHO anyone really serious about vinyl will be using an MC in 90%+ of cases. A MM only version is also easily possible and I am in fact about to start on one, based on the hybrid mentioned above, as a custom build for a customer....

The other big problem is that without negative feedback to force all the circuitry to behave as intended, the parameters of the individual transistors become crucial to proper operation. This means painstaking matching of the devices, as all Paradise builders will attest to! This is very time consuming and fiddly plus it results in maybe 70% of transistors bought in being rejected... all very expensive if doing it commercially.

A major part of the design of my own TPS is to try and minimise the amount of devices needing careful matching.... It's down to 4 per channel that really need to be very well matched and a few more that it helps if they are somewhere near in this present incarnation. Only 1 per channel in the hybrid TPS versions...

I can't see any point in photos of some circuit boards... which for obvious reasons can't be close up and HD.... :?

Thanks for the offer of help with casework! :D I could certainly use the help. It will kind of need to be soon though I'm afraid as I'm about at the stage where it's going to have to go out for demo to any interested parties tout sweet in order to hopefully get some orders and amortise all the R & D that has gone into this so far... even if it has to be built into the carcass of a rusty old battery charger or whatever to get it out there...

YNWaN
Site Admin
Posts: 3023
Joined: Wed Apr 27, 2011 2:02 pm

Re: Arkless Electronics Transconductance phono stage

Post by YNWaN » Sat Oct 21, 2017 3:32 pm

What size case would be needed?
Mark

Arkless Electronics
Posts: 64
Joined: Mon Oct 16, 2017 5:54 pm

Re: Arkless Electronics Transconductance phono stage

Post by Arkless Electronics » Sat Oct 21, 2017 3:48 pm

sq225917 wrote:
Fri Oct 20, 2017 10:42 pm
jez, get it ship shape and I'll try and get over next weekend. I've just finished my own Paradise PSU so we can give it a good compare and contrast. What mains voltage do you have, i'll trim the psu to suit.
Great. I could do with the space from you collecting your Muse for a start... especially as I had an unsolicited delivery of 5 boxes of gear from Richer Sounds head office yesterday which I know nothing about! Not even a covering letter with it...

It will be very interesting to compare it at its current revision level against your Paradise as well!
A big problem at the moment is that having blown a tweeter on one of my beloved KEF R 105/3's I'm reduced to using Spendor BCII's for my listening tests and the KEF's are so much better than the Spendors that even "the worlds best phono stage" through the Spendors would not match even the Arkless 640P through the KEF's! :roll:
The Spendors are at least pretty revealing, even if the overall SQ is not in the same league as the KEF's, and seem to be well capable of showing that a particular change I make is for the better or not. BUT specific favourite test tracks don't sound as good through the Spendors with this TPS as known very good phono stages of my own design obviously did through the KEF's, so it's impossible, or at least very difficult, for me to give an opinion as to where in the pantheon of phono stages this TPS actually sits in its present form...

Even up here in the frozen north mains is a standard 240V 50Hz :mrgreen: I don't often measure it but it's usually within 5V of 240 on the occasions I have checked.

Arkless Electronics
Posts: 64
Joined: Mon Oct 16, 2017 5:54 pm

Re: Arkless Electronics Transconductance phono stage

Post by Arkless Electronics » Sat Oct 21, 2017 5:06 pm

YNWaN wrote:
Sat Oct 21, 2017 3:32 pm
What size case would be needed?
It's a two boxer (as all phono stages really should be to keep the nasty noisy mains transformer away from sensitive circuitry).

Phono stage will need to be 250 x 260 x 70 and PSU box 250 x 150 x 110 (it's an 80 VA ish by the look of it twin "C" core transformer so a big un by phono stage standards.... it was to hand, good quality and of suitable voltage). Those are not absolute limiting dimensions but allow 1/2" - 1" wriggle room... The actual TPS has grown in size already due to extra off-board PSU decoupling caps for example.
The TPS box needs to be as close to hermetically sealed as possible... certainly no ventilation slots or anything like that!

Post Reply