John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Siemens Styroflex too, when was the last time you saw leads that thick on a polystyrene cap?

Red ring means it's a 160V part, 4640pF implies it has 0.6mm Ø leads.
Actually, the low budget of the Siemens KS series, polystyrene & aluminum foil.

Siemens boxed type B31531 is polystyrene & tin foil, 63V parts, lead diameter 0.6-0.8mm.
Both are good for 200.000hrs minimum, but the boxed ones have ~50% better numbers (tan δ, TK, dropout number, etc.), at 4 times the cost for the value in the picture.
(under 220pF, 10 times as much, >$5/pc in the early '90s)

So it's all relative. :clown:
 
Last edited:
diyAudio Member RIP
Joined 2005
Mr Wood,

you seem to have mentioned between a thousand and +5K of Mr Wurcer's favorite JFET in parallel.

:)clown:)
That's in the ballpark. I think in bed unable to sleep this AM, I figured about 800 BF862, assuming you could use 150mW of the phantom power, say at 3V, for the FETs. This would be to achieve around the thermal noise of a one ohm ribbon at 300 K. The Brownian and thermomechanical noises are likely higher, let alone normal studio ambients, and iirc Scott mentioned that a considerably noisier preamp still worked quite acceptably in real-world direct ribbon use.

But it's still an interesting stunt, a cunning stunt as the risque joke goes. Note that this approach only works for FETs and does not work for bipolars, which would otherwise probably be the devices of choice (some of the 2 ohm rbb' ones run at healthy currents) where the mic could have a separate power supply.

Oddly, many years ago I did some experiments, on 2SK170 equivalents and a couple of the big Crystalonics parts, in response to a challenge from Pease, who'd been busy in print emphatically disbelieving one of his favorite whipping boys, Keats Pullen. However I did not measure noise at the time, just followed the transconductance versus drain current down to very low levels (nanoamps). As the data vindicated Pullen, Bob was evidently disgustipated worse than Popeye and almost completely ignored the work, mentioning it briefly in a paragraph of a "cleanup detail" column and getting most of the details wrong. But we shall say nothing more ill of the dead.

However, it is possible that the noise performance deteriorates faster going to low currents than first-order theory predicts. A couple of people I recall said they were seeing some evidence of this, albeit not perhaps conclusive.
 
My recommendation for a future preamp:

A beautiful piece of hardware, almost a shame to put an opaque cover on it. With components like that, I'd build shelving tipped forward 20 or 30 degrees to show off the guts..a geek thing of course..:D

The grounding scheme however, is terrible. Any time varying magnetic field trapped between any left channel IC and any right channel IC will push loop currents through the green ground wires. This couples to the signal wires.

Coax shield does not shield magnetic. This design forces the customer to tightly bundle ALL input cables together. Otherwise, there may be hum/noise issues.

jn
 
Member
Joined 2004
Paid Member
A beautiful piece of hardware, almost a shame to put an opaque cover on it. With components like that, I'd build shelving tipped forward 20 or 30 degrees to show off the guts..a geek thing of course..:D

The grounding scheme however, is terrible. Any time varying magnetic field trapped between any left channel IC and any right channel IC will push loop currents through the green ground wires. This couples to the signal wires.

Coax shield does not shield magnetic. This design forces the customer to tightly bundle ALL input cables together. Otherwise, there may be hum/noise issues.

jn

It's crying out for a PCB across the connector array with all the switching/transformers mounted on it. Much better control of everything but no magical wire to make the sound better than it actually is. And an empty machined aluminium box (it is now but the wire makes it look less empty).
 
strong in innovating.

Also the first commercial echo sounder for fishing vessels (Fischlupe)
The echo-sounder was also a German invention, 1918.

(Elac/Electroacustic started out as Signalgesellschaft, a company that developed sound apparatuses for submarines)
 

Attachments

  • Fishy Signalgesellschaft.jpg
    Fishy Signalgesellschaft.jpg
    26 KB · Views: 248
I worked on such a design for a while, mostly as a stunt, until distracted by more immediately remunerative pursuits. A lot of the work was on the efficient but low noise power conversion.

The well-regarded Royer R-122 still uses a matching transformer, but they may not need the whole 1:50 stepup. Pretty certain they didn't include DC-to-DC conversion either, but still a solid and practical commercial compromise. (That's *not* a bad word.)

I once calculated the number of 12AX7 triodes it would take for 3dB noise figure from a nominally 3 Ohm Coles. IIRC it was 853, but can't remember if that was triodes or whole (two triodes each) 12AX7's. And ignored excess noise. And ignored the wire resistance from the ribbon to the electronics. Yikes!

Thanks, as always,
Chris
 
well OK i'm not surprised at the freyed blue sleeving now, the website is loud and covered in that blue. I particularly love this one

An externally hosted image should be here but it was not working when we last tested it.


love the heatsinks...

I guess the cases are designed to make them look more substantial, or match other components, but its still pretty funny, I wonder if they are extra
 
Last edited:
Status
Not open for further replies.