P2P Redd 47 - a few questions

[Jean Clochet]

Post removed.

 

[abbey road d enfer]

According to this link
http://forums.vintageamps.com/viewtopic.php?f=13&t=74430
it seems almost impossible to relate performance with brand/denomination
CV4085 and Z729 were just qualified as long-life, not implying any deviation from the EF86 specs. Any superior performance, ascertained or alleged, must have been incidental.

 

[earthsled]

I thought it interesting that "skirted valve holders" were mentioned as a means to reduce hum in the guidelines Abbey recently posted. Would this hold true for PCB designs where the tubes are board-mounted inside a chassis? If so, would grounding the skirt/shield to signal ground be equivalent to using the chassis ground?

I also wonder how ceramic sockets compare to PTFE with regards to hum.

Thanks!

 

[abbey road d enfer]

Yes, somewhere I believe I mentioned briefly that implementation is as important as comonent selection in that respect. You need to shield as much as possible and ground pour.

 

[earthsled]

Yes, somewhere I believe I mentioned briefly that implementation is as important as comonent selection in that respect. You need to shield as much as possible and ground pour.

Okay, shielding the tubes is definitely a good idea. Does it matter if the shield is tied to chassis ground or signal ground in a PCB situation? Also, the attached guidelines suggest PTFE sockets are superior to nylon loaded phenolic. I have a choice of micalex or ceramic -- is either material better for hum reduction?

Thanks!

 

[abbey road d enfer]

I would think ceramic.

 

[earthsled]

Great, thanks!

Sorry, this is a bit off the topic...

I've been corresponding with Kevin at K&K Audio about using the Lundahl LL1576 input transformer for this circuit. He tells me that it should work without needing a Zobel on the secondary (eliminating C13 and R1). The LL1576 is comparable in price to the CineMag CMMI-7C, but it is a PCB-mount config. (This is why I've been asking about tube socket grounding on the PCB).

My question is: If there is no Zobel, will I need to parallel a resistance with the 1M grid leak to reduce the reflected impedance?

 

[abbey road d enfer]

I've been corresponding with Kevin at K&K Audio about using the Lundahl LL1576 input transformer for this circuit. He tells me that it should work without needing a Zobel on the secondary (eliminating C13 and R1). The LL1576 is comparable in price to the CineMag CMMI-7C, but it is a PCB-mount config. (This is why I've been asking about tube socket grounding on the PCB).

My question is: If there is no Zobel, will I need to parallel a resistance with the 1M grid leak to reduce the reflected impedance?

No, you don't. Many vintage preamps (RCA in particular) had the same topology, with the secondary directly connected to the grid.

 

[CJ]

the GEC tube might have been a 6267. kind of a mil variant of the ef86, and prefered for vox ac30 especially if you get the amperex holland version.
lower the screen volts for long life without noise.

dc is great for heaters, mucho hum reduction, provided that the rest of the system can take advantage of this (ie: it is quiet)

regulators are a bad idea for heater supplies, they just seem to get nuked, too much inrush current when the tubes are cold, maybe use the pass transistor to compensate,

just a simple heavy duty bridge rect with a pi filter with mega C, and you are good to go.

some folks say dc is bad because of cathode stripping, but who wants to live that long?

 

[Jean Clochet]

Post removed.

 

[earthsled]

I'm wondering about how much distance is needed between the power transformer and other components in the circuit. I would assume the input transformer and V1 should have the most distance from the PT, whereas V2 and the output transformer could stand to be closer.

Attached is a first draft of my 2-channel layout. The rack chassis is 8 inches deep. Tubes and input transformers are currently mounted on a PCB, and the power transformer mounts to the rear panel of the chassis. The tubes are positioned to be under the ventilation in the top lid.

As an alternative, I've been toying with the idea of using an external box for the power transformer, but I'm leery of having high voltage on a connecting cable.

Any comments or suggestions?

Thanks!

 

[abbey road d enfer]

Transformers are sensitive to orientation, you may have to experiment before layout.

 

[earthsled]

Modern production tube shields and skirts seem to be made from aluminum or some non-ferris alloy. Will these aid in reducing hum at all?

 

[abbey road d enfer]

Yes, because what you want from them is electrostatic shielding.
Using magnetic materials may concentrate or deviate in an undesirable manner electromagnetic interference.
Magnetic shielding should be handled by specific screens that can be moved until an optimum position can be found.

 

[letterbeacon]

I've heard steel is good good for magnetic shielding, but is it worth splashing out on some mumetal sheets?  They're pretty expensive on eBay, but if they work they might be worth it...

http://www.ebay.co.uk/itm/Ultraperm-MUMetal-Shielding-Sheet-0-15mm-31-5cm-x-19-5-/230446633571?pt=LH_DefaultDomain_0&hash=item35a7b09263#ht_1454wt_907

 

[abbey road d enfer]

Just google "ultraperm".
You'll have lots of threads about how it's fantastic, and how it's pure B.S.
The truth, as always is in between.
50% of success with that type of material is application.
The material must be cut with as little mechanical stress as possible, so use very sharp scissors, no sharp bends, no holes drilled.
And a little theoretical knowledge about what it does helps too.
Magnetic shielding does not absorb magnetism, it works by reorientating the flux.
And generally, it is an open loop (the adhesive is non-conducting).
That's why you'll see sometimes a complementary copper shield that is generally soldered to ensure electrical continuity, making it a closed loop.
It can be somewhat considered as absorbent; it works by creating a counter-induction that tends to null the incoming flux.
As some member here has into his sig, this sh...t is actually science.

 

[Davo]

I agree with abbey... I've used it a number of times, and it's been good, but it's pretty thin stuff and I usually have to double up on it to get
significant results.  In any case, the same can be had here for a lot less cash:

http://www.goldmine-elec-products.com/prodinfo.asp?number=G16600A

or if you want something that's a bit heavier:

http://www.lessemf.com/mag-shld.html

cheers

 

[letterbeacon]

Very interesting, thank you.

I was thinking of putting a steel panel with the Ultraperm stuck to it in between the PT and the output tranny I have in my preamp.  The idea being the Ultraperm will provide most of the shielding and the steel plate hopefully providing a bit more.

Abbey - would a copper plate be better?  Solder the Ultraperm to the copper sheet and then bolt the copper sheet to the chassis?

Thanks for the links Davo but unfortunately Goldmine don't ship to the UK.  This place does though, apparently:  http://www.diyhifisupply.com/node/695

 

[abbey road d enfer]

Very interesting, thank you.

I was thinking of putting a steel panel with the Ultraperm stuck to it in between the PT and the output tranny I have in my preamp.  The idea being the Ultraperm will provide most of the shielding and the steel plate hopefully providing a bit more.

Not the most efficient way. Air/metal/air /metal/air is better.

Abbey - would a copper plate be better?

Not better. Different.

Solder the Ultraperm to the copper sheet and then bolt the copper sheet to the chassis?

Don't do that. Heat destroys the thermal treatment. That's why it's adhesive.

 

[letterbeacon]

Solder the Ultraperm to the copper sheet and then bolt the copper sheet to the chassis?

Don't do that. Heat destroys the thermal treatment. That's why it's adhesive.

Ah I see, I think I misunderstood you there.  Thanks!