Latest Design - Tube channel strip

[BluegrassDan]

Hmmm. Maybe I need to look into subbing the 12SN7 for 12AU7. Does anyone have documentation on how to change values of neighboring component?

 

[mjrippe]

That is not a bad contender. Definitely an improvement on the 12AU7 but its is a proper big power tube. Much taller than the AU7 and takes twice the heater current. I used it once in a headphone amp.

Cheers

Ian

Yes, however same heater current and overall height as 12SN7 - but in a 9 pin miniature base.  Also 12BH7 is in current production.

 

[ruffrecords]

Yes, however same heater current and overall height as 12SN7 - but in a 9 pin miniature base.  Also 12BH7 is in current production.

Absolutely. I just wondered if either will fit in the OPs enclosure.

Cheers

Ian

 

[BluegrassDan]

I think I can fit it in.

 

[BluegrassDan]

Low cut test sweeps seem pretty good - at least within a few dB of where they should be, theoretically.

 

[alexc]

I like the jj5751 for a first stage, high gain role - in my measurements, it's been the quietest, beating out the next best -  eh12ay7 and eh6922.

The jj5751 has a less ragged harmonic spectra than run of the mill ax7 - quite similar to the ay7 and 6922.  ie. sparse harmonics, nearly all h2 and with v.low hum and so a v.good signal-hum ratio - where it edges out the competition.

I do think the au7 and similar, along with the av7 types sound quite good but do measure a little noisier and with higher thd.  They can be useful when you want a higher h3 in the mix.

The jj5751 + eh6922 is a great combo    my favs  where I can't use 6SL7/6SN7.

 

[merlin]

Almost any other double triode will be better than that.

That hasn't been my experience. Apart from a small number of special purpose triodes like the ECC88, ECC83, 6SN7 and possibly the 12AY7, most triodes comes out about the same or worse than the 12AU7.  1% at 10Vrms is what I consider average for a small signal triode. :

 

[ruffrecords]

That hasn't been my experience. Apart from a small number of special purpose triodes like the ECC88, ECC83, 6SN7 and possibly the 12AY7, most triodes comes out about the same or worse than the 12AU7.  1% at 10Vrms is what I consider average for a small signal triode. :

I am puzzled. The most commonly used (double) small signal triodes) in audio are ECC81/3/8, 12AY7, 6SN7, 6CG7 and possibly the 12BH7 all of which beat the 12AU7. The only other commonly used small signal triode is the Russian 6N1P and that is almost as bad as the 12AU7. Which other triodes were you considering?

Cheers

Ian

 

[BluegrassDan]

Which of those tubes is most suitable for eq gain makeup V1 (which will be followed by a cathode follower)? The 12AY7 is really too much gain.

 

[ruffrecords]

Which of those tubes is most suitable for eq gain makeup V1 (which will be followed by a cathode follower)? The 12AY7 is really too much gain.

Depends on several factors including how much loss there is in the EQ (assuming it is passive) and what loading the EQ expects.

What is the cathode follower for?

Cheers

Ian

 

[alexc]

Lower in gain than the ay7 means av7 types or 6922 types - lower gain than those , is the au7 and similar.

The differences in gain aren't all that much really - a couple of dB or so. 

I would have thought you want quite a lot of gain in an eq make up stage. You can always use some local nfb or even padding the input/output if really necessary to lower gain.

 

[merlin]

I am puzzled. The most commonly used (double) small signal triodes) in audio are ECC81/3/8, 12AY7, 6SN7, 6CG7 and possibly the 12BH7 all of which beat the 12AU7. The only other commonly used small signal triode is the Russian 6N1P and that is almost as bad as the 12AU7. Which other triodes were you considering?

Well, they're commonly used for audio because they fall on the 'better' side of average, i.e. they're unusually good! They've floated to the top of the  pond of dozens of TV/radio double triodes. Methinks you've been spoiled by years of using the best types! 
(The exception is the ECC81, though, which is worse than the 12AU7 at about 1.5% THD @ 10V into a resistive load, or  0.4% with a CCS.)

 

[ruffrecords]

Well, they're commonly used for audio because they fall on the 'better' side of average, i.e. they're unusually good! They've floated to the top of the  pond of dozens of TV/radio double triodes. Methinks you've been spoiled by years of using the best types! 
(The exception is the ECC81, though, which is worse than the 12AU7 at about 1.5% THD @ 10V into a resistive load, or  0.4% with a CCS.)

Which makes me wonder if there are any other gems that have been missed.

Edit: According to the National Valve Museum the ECC82 and ECC83 were originally designed for audio applications. Most of the other small B9A double triodes were designed for VHF front ends so good audio performance would be serendipity.

Cheers

Ian

 

[merlin]

The only other commonly used small signal triode is the Russian 6N1P and that is almost as bad as the 12AU7.

The 6N1P almost as bad? Heaven forfend! It's pretty much as good as a 12AX7 into a resistive load,* with about 0.6% THD @ 10vrms. I knew I'd posted it somewhere:
http://www.diyaudio.com/forums/tubes-valves/274670-list-tube-noise-measurements-please-nominate-lowest-noise-tubes-27.html#post4678954

*I don't have data for 6N1P with CCS load atm

 

[ruffrecords]

The 6N1P almost as bad? Heaven forfend! It's pretty much as good as a 12AX7 into a resistive load,* with about 0.6% THD @ 10vrms. I knew I'd posted it somewhere:
http://www.diyaudio.com/forums/tubes-valves/274670-list-tube-noise-measurements-please-nominate-lowest-noise-tubes-27.html#post4678954

*I don't have data for 6N1P with CCS load atm

You must have tested some really bad 12AX7s. The ones I tested were a lot better than that.

Cheers

Ian

 

[BluegrassDan]

I messed around yesterday with 12AU7 versus 12AY7 for the eq circuit makeup gain stage. The distortion figures were negligible compared to the problems associated with the increased gain of the 12AY7. I’m gonna roll with the 12AU7 for a while and see how it goes.

Here is a video of my experiment: https://youtu.be/aS99VQ-9sV8

 

[ruffrecords]

Love the video and the track!

Cheers

Ian

 

[BluegrassDan]

I think the best way to handle makeup gain for both the compressor (first in the chain) and the EQ is to use the EQ makeup gain amp for overall makeup gain after both sections (with a stepped gain switch). Is it wasteful to drop the gain of the compressor with a fixed resistor rather than a pot (in other words, set the pot of the LA2A to a reasonable level, then replace with a fixed resistor)?

Is the compressor’s makeup gain circuit even necessary at all?

 

[ruffrecords]

I think the best way to handle makeup gain for both the compressor (first in the chain) and the EQ is to use the EQ makeup gain amp for overall makeup gain after both sections (with a stepped gain switch). Is it wasteful to drop the gain of the compressor with a fixed resistor rather than a pot (in other words, set the pot of the LA2A to a reasonable level, then replace with a fixed resistor)?

Is the compressor’s makeup gain circuit even necessary at all?

Since the compressor reduces the signal level when it compresses, it is necessary to add gain to restore the signal to the original level so the compressors makeup gain is necessary.

The nominal gain of the LA2A gain make up amp is well over 30dB (set by the negative feedback network). However,if the maximum gain reduction is only 20dB then there is little point in having more than 20dB of make up gain so you could use a much simpler circuit (half a 12AU7 for example). You should keep the compressor gain control just before this amp and feed it into the EQ. Only in this way can you turn down the level going into the compressor make up amp so it does not overload when there is no compression (you effectively have 20dB loss in the pot followed by 20dB gain make up)

It is arguable that some of the tone of the EQ inductors comes from driving them at line levels so that would also argue in favour of having the compressor gain make up before the EQ.

Cheers

Ian

 

[BluegrassDan]

Silly question. Could each half of a 12ay7 (or 12au7) be used - one side for compressor makeup and the other half for eq makeup?