Rouge Un - a red preamp featuring Lundahl , 5534 & pumpkin output

[KrIVIUM2323]

Yes my description was confusing, let me rephrase it: on one deck you switch attenuation before the opamp for the 5 first positions then for 6 to 11 you change resistor value around the opamp ( feedback loop) on the second deck ( and bypass any attenuation happening before the opamp), the last 12th position is unused/blank.

It's unusual approach to use attenuation before entering opamp. I remember many old time members being sckeptical about this as it have a potential 'to suck sound' ( whatever it means) iirc. Never been bothered by this in use with the 430mk2 i had access to in studio, so i don't know if concerns were legitimate. Neve did things his own way anyway and he probably have some very valid reason to do so...

I think the switch needs to be mbb but it's been a while since i had a 430 opened in front of me so i might be wrong about it?

 

[thomasdf]

For ultimate p*rn I was mostly concerned with finding originals TDA1034N for the first stage

Re the switch, I’ll prototype and report back one of these days.
Doesn’t seem strange to attenuate before gain, it’s commonly done with tube designs for instance, and Rupert’s early solid state designs are inspired by / take roots in tube design : transformer isolation and general topology / design blocks

 

[laitue5]

Hello, I'm in for two pumpkins.

 

[KrIVIUM2323]

For ultimate p*rn I was mostly concerned with finding originals TDA1034N for the first stage

Re the switch, I’ll prototype and report back one of these days.
Doesn’t seem strange to attenuate before gain, it’s commonly done with tube designs for instance, and Rupert’s early solid state designs are inspired by / take roots in tube design : transformer isolation and general topology / design blocks

Hi Thomas, why 1034N? This is what was used into the desk? I was pretty sure it was 553x, but i might be wrong.

Tube/opamp topology: tube are high input Z, 553X are not. It make a difference in that case imho. If you take a look at other 'classic' 5534 with input transfo console you won't find any with attenuation implemented this way ( no Ssl, no Ams/Neve, so far no Amek either ( afaik),...).
If you know of other circuit sharing this kind of topology with opamps then i'm all hears.

 

[tk@halmi]

Re the switch, I’ll prototype and report back one of these days.

The Grayhill 71 series has nice PCB versions and are well regarded. These days the 2x12 configurations are rarely stocked, but often they can be ordered in quantities of 25 or more.

 

[thomasdf]

Hi Thomas, why 1034N? This is what was used into the desk? I was pretty sure it was 553x, but i might be wrong.

Tube/opamp topology: tube are high input Z, 553X are not. It make a difference in that case imho. If you take a look at other 'classic' 5534 with input transfo console you won't find any with attenuation implemented this way ( no Ssl, no Ams/Neve, so far no Amek either ( afaik),...).
If you know of other circuit sharing this kind of topology with opamps then i'm all hears.

True !
I was thinking of early transistors designs with in & out transformers and fixed gain from the likes of RCA, Langevin & such, which are based on tube designs blocks, and where input attenuation before gain is very common.
Re the TDA1034N : word on the street is it based on a Rupert Neve opamp design. It then evolved into the NE5534 and they can be swapped, though they have different cases
Some people whom ears I trust told me the TDA 1034 sounded "better" or more "interesting" especially at high gain settings, so I'm curious to hear one some day and I am seriously thinking of allowing some space on the PCB to use either a 1034 or a 5534, kind of paying tribute to the man behind the original design I am a sentimental guy, I know, I know..

 

[thomasdf]

The Grayhill 71 series has nice PCB versions and are well regarded. These days the 2x12 configurations are rarely stocked, but often they can be ordered in quantities of 25 or more.

That's good to know... maybe we'll get into another feeler or even a group buy then haha

 

[KrIVIUM2323]

Thank you for clarifying your point Thomas. Yeah early transistors design shared some attribute with tube, first one being jfet behaving more or less the same regarding impedance ( and biasing, and,... ).
Then bjt appeared and things changed a ( tiny) bit. Whatever, great things from all this technology can be expected when designed by talented peoples.

I have heard same thing about R.Neve involvment into what ended as 5534. It could explain why it is such 'bulls eye' for high quality audio duty despite it's age.
There is not many 'better' audio opamp availlable even this days. Some 'niche' or dedicated to 'special' features but overall not many more are 'better'... But i disgress...
Do you have example of gear which used the Tda 1034? I think i never seen one used despite i known about it's existence.
Good idea to give choice between both on your pcb imho.

 

[JMan]

A small update about the output transformer. I have sent my NOS pumpkin to Moby. This transformer will be sacrificial, as complete analysis requires that it must be torn down to ascertain important aspects of the design such as the core type.

I want to continue to encourage folks who are interested to chime in. As of the most recent count, we are past halfway to the minimum quantity for Moby to be able to produce a run for us. This could be a great opportunity for us DIYers to get ahold of something that, even in this age of abundance in terms of repros of classic components, is still pretty much impossible to get.

 

[thelivingroom]

That's good to know... maybe we'll get into another feeler or even a group buy then haha

If I keep coming across more projects on this site and adding them to my backlog I may end up meeting the 25 minimum on my own haha.

 

[Matt Syson]

The TDA1034 (Philips) was used in some mixers, notably Neve and audix in 1979 or thereabouts and were around 5 Pounds (sterling) apiece so only used in circuit positions that could justify the cost (around 3 hours of engineer time!). Then the design of the TDA was 'released' for other manufacturers and the 5534 came about, with minor improvements (stability perhaps?) and noise performance.
Sometimes it is VERY difficult to improve on 'the best for a given application' and the universal 'perfect OP amp has never appeared although specialist designs have got very good and for example the basic op amp has been expanded to include popular circuit blocks like the SSM 2015/6/7/9 'mic/transducer amplifiers.

 

[thomasdf]

I wouldnt be surprised if some Philips / Polygram consoles used them. It cannot be reduced / narrowed down to that single component but boy they do sound absolutely gorgeous.

 

[inf0]

In for 2 please

 

[thomasdf]

Yeah, keep it coming !

 

[Dylan W]

Do we have any sense of cost? If they’re toroids with an ordinary core, I would think they should not be too expensive.

 

[JMan]

As I mentioned back in post #60, we don't currently have an estimate of cost. I have a transformer en route to Moby to analyze, but it may take some time to reach him and then I don't know what kind of timeframe we're looking at for teardown, all of which needs to happen before we'll know an approximate cost.

I, too, am hoping this will prove to be rather affordable, but for now we don't know.

 

[KrIVIUM2323]

Affordable is a rather relative term...
I mean a Lamborghini Gallardo can be bought used for only 90.000$ which is affordable versus a new Huracan STO at 400.000+ $...
I would need 100 of my own used Dacia Sandero value for an STO, which is a bargain imv ( i don't care for the sex appeal with cars obviously).
I hope we would be in Dacia range rather than Lambo for this transformers! After all they are Pumpkins... Lol.

 

[thomasdf]

Also, we are not collecting money, just seeing if there is interest for such a transformer repro.
If there's interest, then Moby will tell us how he sees things construction wise, cost wise.
And then he'll come up with a price and people will be free to buy or not.

Obviously we all hope it will be cheap, heh. @Moby will tell.
And you can all thank @JMan who bought an original unit for it to be studied/destructed !

 

[De-El]

Depending on final cost I would be intersted in 2-4 units.

D

 

[tk@halmi]

I wonder how "authentic" these TDA1034s could be . It has an NXP label but it is not listed on the NXP website.

https://www.aliexpress.us/item/3256806792602221.html?spm=a2g0o.productlist.main.1.584fg9KLg9KLVf&algo_pvid=b0aabd05-8787-421b-ae6b-0a4fa5b8f13f&algo_exp_id=b0aabd05-8787-421b-ae6b-0a4fa5b8f13f-0&pdp_npi=4@dis!USD!1.40!1.40!!!10.00!10.00!@2101ea7117310926479382581ec51c!12000038922586390!sea!US!2212446183!X&curPageLogUid=D8Wq7haZuxOl&utparam-url=scene:search|query_from: