[Finch&Music]

I take the risk of asking a rather stupid question, but ................

At a german site I see divices that can let you see the amount of bias voltage. See for yourself at Thel.de

Go to their Tube sections, more specific the lcd instrument to give information on tube performance for biasing voltage.

Those divices are in my opinion rather pricey, but could give you on the outside of your case severly wanted information about some aspects of the performance of the tubes in your Millet Amp.

That company use a lcd to let you see on the front of the amp-case what bias is actually performing.

May be such a divice can be combined with an off board potmeter mounted on front of the case (instead of the on board pot on the Millet Amp).

In that way we can always from the outside finetune the bias setting. In that way the biassetting can be done instanteniously and after the temperature inside the case has stabalised.

I hope this make any sense.

May be there are more cost effective solutions to make those bias figures actually visiable and adjustable from the outside.

I am interested in your reactions on this topic.

Pieter Vink.

 

[JWFokker]

Quote:

Originally Posted by drewd Yes, it was helpful for my Grados and DT-880s. -Drew

What benefit did this have? Being a fan of Grados, this interests me.

 

[n_maher]

I'm sure my statement will require a bit of correction but my understanding is that stacking the buffers (running them in parallel) increases the current output capability of the amp, which is exactly what the higher end Grados need to shine.

HTH,

Nate

 

[drewd]

Quote:

Originally Posted by JWFokker What benefit did this have? Being a fan of Grados, this interests me.

It cleared up a bit of distortion that I was hearing at higher volumes with dynamically active music, especially in the low end.

Adding a STEPS with a 1.5A regulator helped a lot, too.

-Drew

 

[JWFokker]

I've seen them stacked as much as four high on Pimetas. Any recommendation for the Millett Hybrid?

 

[drewd]

Quote:

Originally Posted by JWFokker I've seen them stacked as much as four high on Pimetas. Any recommendation for the Millett Hybrid?

Two works just fine for me. As with all things audio, let your ear be your guide.

-Drew

 

[amb]

Speaking of the BUF634Ps, I have an idea that I might try when I build my Millett Hybrid.

Since the buffer here runs open loop, and the BUF634 is itself an open loop buffer, I wonder if there might be a benefit to using something that's closed loop instead. Also, unlike in the Pimeta or MINT, the buffer here is not wrapped inside a global negative feedback loop.

The BUF634P happens to have the same pin-out arrangement as a DIP-8 single-channel opamp, except that the BUF634P has no connection to pin 2 (opamp inverting input), and pin 1 is used for bandwidth/quiescent control (not used on most single opamps). I think it might be interesting to substitute a TI/Burr-Brown OPA551PA high-current opamp in place of the BUF634, and install a small wire jumper between pins 6 and 2 at the DIP-8 socket. This makes the OPA551 operate as a closed-loop voltage follower. The OPA551 is rated for 200mA continuous output current, which should be high enough for most headphones.

Granted the OPA551 doesn't have quite as high a slew rate as the BUF634, but it's not really that slow and it's being driven by a tube which is not known for ultra high-speeds anyway, so that is not an issue. I wonder if there might be an appreciable difference in sound, possibly an improvement.

Btw, the OPA551 is cheaper than the BUF634 (Digikey prices).

Another chip that is a possible candidate for this trial is the Analog Devices AD811AN. It has very high slew rate but lower output current (100mA max), and more expensive.

Just food for thought for now...

 

[Emon]

I'm still waiting on those discrete buffers. Methinks it'll be a huge jump in sound quality...man I love this amp as it is, I can't imagine it with a discrete output stage!

 

[rreynol]

C'mon people, let's see some more builds. There were what, 200 board ordered?

Let's see em.

 

[pabbi1]

Or some (additional) listening impressions?

 

[rreynol]

I like it. In addition, I do agree with the other observations concerning 12AE6A vs 12FK6 tubes (the latter providing more detail and sounding better). Once my parts arrive for amp #2, I'll have some more pictures to share.

 

[pabbi1]

OK, I'm convinced - I like the 12AE6A pretty well, but wanted to burn in the amp first before switching over to the 12FK6. Any observations about the tubes having a "wake up / warm up" period (after the initial 30 minute warming)?

And, is there a significant difference in the 'bias'?

Edit:

My 16v bias for the 12AE6A was an initial 9.9v bias with the 12FK6, given about -1v for them being not warmed up.

And, I must agree: the 12FK6 are the better tubes. In my case the 12AE6A are RCA and the 12FK6 are Tung Sol.

 

[Secret Squirrel]

Ok, I tried stacking the buffers a few minutes ago. My lower impedance Ath-A900’s (vs. HD-580’s) seem to react to this change positively. I need more time comparing this change to my single buffer unit.

 

[SHLim]

A question about the tube socket. I know I need to solder all the socket feet, but do I need to solder the center metal piece on the socket to the pcb?

TIA,
Sam

 

[n_maher]

Quote:

Originally Posted by SHLim A question about the tube socket. I know I need to solder all the socket feet, but do I need to solder the center metal piece on the socket to the pcb? TIA, Sam

Nope, you may notice that the sockets that some of us use don't have a center piece at all.

Nate