Rouge Un - my attempt at a red Lundahl + 5534 preamp

[a soBer Newt]

A hopefully quick hijacking, I have been curious as to the cap (C10) in-between the two 16K in the feed back path and what its trying compensate.

 

[thomasdf]

I think it need more than just passing a signal.

I am inclined to believe what Lundahl says but since you have the 1539 already there is no harm trying out.

The original transformer is not that expensive, I am wondering if Carnhill would sell directly or it's a proprietary design.

If they don't you should just buy a spare one from the link above and reverse engineer it, I know several toroidal transformer makers that could make a small batch if we can organise a group buy.

I have even found another 1539 in my stash haha, maybe they made a baby when I played that Marvin Gaye record the other night ...
The thing is, I'd like to use them. The Carnhill pumpkin is obviously tempting... but for me it is buying more stuff when I want to clear the stash. The thing to take into consideration too, is that the Carnhill tx makes it a more complicated project, mechanically speaking. Lundahl is on the PCB, end of story. Carnhill means the PCB is not full size 500, so both the TX and the PCB need to be fixed to the L bracket, and that leaves less room for the PCB surface, which means a more cramped layout.

Do you have any idea / clue what would be the cost of the toroid output rebuilt ? At this point if there is interested for the project, I am open to buy and rev eng a pumpkin, and then talking about it and taking a somewhat collegial decision.
What was my project to clean the shelves and me sharing the design can be bent into a group reflexion / conception and buy.

 

[amplexus]

Without speaking to this design specifically- I am somewhat familiar with the 1539 and I know that it is an odd duck in terms of its connections and windings. My Amek BC3 uses them as mic input, line input, and line output transformers.

I don’t know the specifics but I know that the people I have come across who know about the dark arts of transformer design have said that it really is a unit designed to be used exactly as the datasheet application note specifies to get the performance noted. Lundahl seems to be very… rigid… in their customer service and support in my experience so they are likely saying “the 1539 will not work. Use this instead” because they are concerned about the specifics and absolute quality of performance is my bet.

I say that because in the BC3 they are connected quite conventionally and they work quite well and that preamp circut (which is similar to the one discussed here) is well regarded. So I don’t see why they can’t be made to work in this design

 

[thomasdf]

Now you make me want to have a look at the BC3 output schematics haha ! Thanks for sharing the info !

I totally respect Lundahl's high standards, and never meant to discuss their posture or even say they are wrong, by the way.
But I happen to have three LL1539 on my hands and I'd like to put them to use

I'll see if there's interested for a 1:1 circuit and a Carnhill Pumpkin repro, or if I readjust my idea and try to get inspiration from the BC3's use of the 1539.

Edit : I found 22 pages of schematics but it is missing the output stage, so if you have that document somewhere @amplexus I am interested. Apparently the same british console genius designed the circuit discussed here, and the Amek BC3 ? I see 5534 are on the signal path ... It seems the designs are somewhat related / in a similar spirit

 

[amplexus]

Now you make me want to have a look at the BC3 output schematics haha ! Thanks for sharing the info !

I totally respect Lundahl's high standards, and never meant to discuss their posture or even say they are wrong, by the way.
But I happen to have three LL1539 on my hands and I'd like to put them to use

I'll see if there's interested for a 1:1 circuit and a Carnhill Pumpkin repro, or if I readjust my idea and try to get inspiration from the BC3's use of the 1539.

Edit : I found 22 pages of schematics but it is missing the output stage, so if you have that document somewhere @amplexus I am interested. Apparently the same british console genius designed the circuit discussed here, and the Amek BC3 ? I see 5534 are on the signal path ... It seems the designs are somewhat related / in a similar spirit

I've got the complete, giant arse freaking binder of em! Quick pics below of the input stage of the preamp modules and output stage of the group/master outs. You can see in both cases, it's just strapped as if it were a single winding per side, with the two windings in series. I would imagine that given Lundahl's hype around how low distortion their datasheet circuit is, that the result of using it as it is here would be slightly higher distortion levels.

And yes, you are correct about the designer of this preamp being one and the same! There are certainly some similarities.

 

[thomasdf]

Thanks a lot for sharing this ! That J5 jumper got me confused, haha, but I understand it's a way of bypassing the output transformer (which was an option, right?). So instead of the AB output stage from the design I shared, here the output is driving with the AT8 Line Driver assembly, and no feedback.
From what I found, the AT8 is dual 5532 + dual 5534 arrangement. It must be able to push a lot of power ?! Make sens to use a 2:1 transformer, rather than a 1:2 transformer with the AB output + feedback.
Maybe that whole AT8 can replace the second IC in my schematic.

But also must sound quite different... I am a bit puzzled !

Lots of great info here :
https://groupdiy.com/threads/amek-at8-hybrid-line-driver-questions.84706/

I'll keep on studying

 

[amplexus]

Yea those jumpers are just to be installed if you didn't install the output trafo. On the input you'd use the common mode choke and resistor arrangement in place of the trafo. They don't make it clear on the schematic but R231/2 and K2 get omitted if you use the 1539. In fact they occupy the exact same real estate on the PCB so they coudln't be installed if the trafo was there. (this is a spare strip, don't worry, all my daily drivers are re-capped )

 

[thomasdf]

Since there has not been any sign of interest (and that's OK ) in pursuing to original design with the Carnhill transformers, I'll keep on working on the projet as being more personal, and decided to try to use an AT8 output to put my LL1539 to use.
I am working on the schematics these days, and hope to share a version pretty soon, to have it double checked and start working on the PCB design !

 

[JMan]

I'd be interested in the original design with the Carnhill pumpkins, but if I'm the only one, I don't suppose it's worth the hassle. How well do the Carnhills fit into the 500 series dimensions? They seem large to me. Maybe they are better suited to a racked version? I'm probably wrong, but anyway, this is a really cool project! Looking forward to seeing more!

 

[thomasdf]

As mentioned earlier the Carnhill means more complicated PCB and metal work, it is quite big. Refer to post #22 and check pictures of the commercial 500 format unit online
Edit : it can be done, though !

 

[thomasdf]

Hi all !
This is were I am right now. A big thanks to @krabbencutter for debunking and drawing the schematics of the Amek AT8 Hybrid Line Driver. It is originally sort of a plugin opamp. I added it as an output stage, replacing the unity gain 5534 and AB output stage from my previous schematic. Too me it looks like a sort of push pull, put maybe the pros con confirm or infirm.

Here the LL1539 is used in a very traditional way, and the output is 1:1 the same as the Amek BC3.

If you smart eagle eyes guys can spot no mistake, I'll move on to the PCB design

Cheers & thanks

 

[mhelin]

Seems to be how the Soundcraft B800 output with an LL1539 option is driven (though it uses the SSM2142 balanced line driver there, the DRV134 could be used these days). Should the output caps be NP - two 1000 uF in series with opposite polarities? Maybe not, in fact the 1000 uF caps are actually already in series in opposite polarities in B800 (the transformer in between of them). You might want to increase the 470uF ones to 1000uF though...

 

[KrIVIUM2323]

I would be interested in pumpkin. Or a recreation. I tried to bought from Carnhill direct some 20 years ago, but got two answers as i insisted after first one: 'this is not a product of our line' followed by 'send us serial number of units ( Focusrite) you own and we might do something'.


Api 2503 or clone should work. Another Lundahl could work too: ll1944.

 

[thomasdf]

Well, the pumpkin is intersting for sure, but it is hard to find ... the website on the previous page does not have them in stock.
2503 does not seem to be a direct drop in replacement ? It is missing a tap on the output ?

@krabbencutter is having trouble with his AT8 rendition so maybe it's a safer bet to stick to the OG design topology.

I guess I'll have unused LL1539 maybe I can sell them to buy something elese

I hate having different options haha

 

[thomasdf]

I had missed an older very interesting thread : https://groupdiy.com/threads/any-one-know-the-focusrite-mic-pre-topology.4485/

 

[KrIVIUM2323]

Yes seen it when i was doing search about the recent thread i posted in drawing board about Neve output/line driver historical evolution.
there is definitely something about the Focusrite sound into this output stage. Some other members told they experienced same difference of sound with eq between transformer/electronic balanced.

I'm convinced it's more into the design rather than the transformer itself.

This 2503 ( afaik it's a clone of original) should be ok as the Cinemag/Reichenbach CMOQ-2:
https://capi-gear.com/catalog/product_info.php?cPath=21_37&products_id=79

 

[thomasdf]

While googling around, I found this : https://zhdesigns.audio/products/z-2503-output-transformer and it got me curious so I bought a pair ... To be continued !

 

[Dylan W]

Well, the pumpkin is intersting for sure, but it is hard to find ... the website on the previous page does not have them in stock.
2503 does not seem to be a direct drop in replacement ? It is missing a tap on the output ?

@krabbencutter is having trouble with his AT8 rendition so maybe it's a safer bet to stick to the OG design topology.

I guess I'll have unused LL1539 maybe I can sell them to buy something elese

I hate having different options haha

2503 has 4 identical windings that are flat out to Mhz range, wire it up as 1:2:1 (with two windings in series for output) as per the original pumpkin and it should work very similarly. As I said before 1539 (which has four 1:1 windings) may also work well, Lundahl won’t sanction it because they probably haven’t tested it in this configuration and they have high quality standards, but it may well work great. You’d just have to try and see.

 

[thomasdf]

Indeed ! I’ll start prototyping one of these days, and will try both the LL1539, as I have three of them sitting unused, and the 2503 too.

 

[Dylan W]

Nice! Btw I believe C14 on your schem should be 47u n/p, in case you missed it