AK Design Collaborative: Tube Preamp/Linestage

dnewma04

The Healer
Design Concept: Collaborative effort to design an excellent linestage with flexible input/output options for members to customize to their needs.

Goals: TBD but leaning towards not using commonly used tubes (12A*7 types), using high quality but non boutique parts.

Gain of 10 or less.

Budget: TBD

Out of scope: At this point, I don't think we'll be making any efforts to create a phono section, but that could be a future project. No tone controls or balance controls, either.

Member Projects:
(Link will be added)

Circuit:
(link will be provided)

BOM:
(link will be provided)
 
Last edited:
Some things that need to be determined will be:

  • Circuit
  • Tubes
  • Gain
  • Budget
  • volume control

I have no opinion on a budget at this point but I'd welcome the group to determine this.

I was initially thinking low mu, but having a bit of gain wouldn't bother me at all if it made the preamp more flexible.

I'm not knowledgeable enough about preamp circuits to offer anything solid in that regard. I'm tempted by doing something unusual like making a pentode preamp because of the flexibility it offers in adjusting gain easily, but if that isn't the way that the group wants to go, things like DHT's or power tubes used as signal tubes is also tempting.

In terms of attenuation, I'm indifferent, but I know that others have strong opinions about the type of attenuation used. I suppose we may want to approach the circuit in a way that reduces the effects of the volume control? I'm ok with any sort of ALPs type potentiometer, as some minimum level.
 
Last edited:
Well, one suggestion that immediately comes to mind- in terms of minimizing volume control influence, and reducing noise- make the design have enough voltage swing, that it will be OK to place the volume control AFTER the gain stage, but before the output buffer. This will provide a constant load to the gain stage, and since a cathode follower has near infinite input impedance, it won't be affected in any meaningful way by being fed from a potentiometer.

By doing this, any noise that is generated in the preamp gain stage, will be attenuated by the volume control. It is usually possible to have an effectively silent background with the volume at low levels, with this type setup.

If you use high-transconductance tubes (with low output impedance), then the normal Alps RK27 100K "Blue Velvet" pot will work fine. It's available from many sources for less than $20, and has proven to be very good, IME.

Regards,
Gordon.
 
I'm a bit intrigued by a few tubes. 4P1L, C3M, and 18040s are all on my radar for this project.
 
I don't want to do tone controls or balance controls, either. Someone interested in balance controls can do a pair of mono attenuators.
 
A couple of other thoughts: Preserve absolute signal polarity. If the preamp does invert polarity it can be corrected at the speaker terminals, but if a good solid design can be made without inverting polarity, it might be something to consider.

Has it been discussed if using silicon is to be allowed or not? For example in the power supply for diode rectification or high voltage regulation, or to aid the audio tubes to deliver their max potential, such as a MOSFET plate load?


Well, one suggestion that immediately comes to mind- in terms of minimizing volume control influence, and reducing noise- make the design have enough voltage swing, that it will be OK to place the volume control AFTER the gain stage, but before the output buffer.

But at the expense of an extra coupling cap, whereas if you put the volume control before the first gain stage you have the option to direct couple the stages if the design would support it, but at the expense of varying input load.
 
A couple of other thoughts: Preserve absolute signal polarity. If the preamp does invert polarity it can be corrected at the speaker terminals, but if a good solid design can be made without inverting polarity, it might be something to consider.

Has it been discussed if using silicon is to be allowed or not? For example in the power supply for diode rectification or high voltage regulation, or to aid the audio tubes to deliver their max potential, such as a MOSFET plate load?

I have no issue with your idea and no expertise in this area. I've heard SS plate loads used to very good effect, I just don't know how to implement them. If someone wanted to take the lead on that and we decided that it was worthwhile to pursue, I'm all for it. In fact, I don't expect to lead any of the design, just leading the effort in getting a thread started and discussion heading in the right direction.



But at the expense of an extra coupling cap, whereas if you put the volume control before the first gain stage you have the option to direct couple the stages if the design would support it, but at the expense of varying input load.

I believe that if we had a pentode line stage, we can minimize any effect from a varying load from a pot up front and also control gain by adjusting the plate load. Something about GM * plate load = gain? Is that correct?
 
But at the expense of an extra coupling cap, whereas if you put the volume control before the first gain stage you have the option to direct couple the stages if the design would support it, but at the expense of varying input load.

Well, direct coupling has its own potential problems too.

If the first stage doesn't bias up at exactly the target point (due to tube variations), then the second stage may be accordingly biased wrong, too. These cascading errors in bias can be more significant in a preamp than in an amp, since the transconductance of the tube somewhat depends on the amount of plate current in the tube. Change that, and you change the overall gain and output impedance, to a degree.

The problem arises, when this happens more on one channel, than the other. We then have a potential situation, where one channel will actually be louder than the other...

A coupling cap isolated design can minimize these bias errors, making the total gain of the preamp more constant over slight tube variations. Unless we're talking about a feedback design (to stabilize gain), having coupling caps between each stage may lead to better gain stability...

Regards,
Gordon.
 
Some people do like tone/balance controls... :)

It would be cool, IMO, to design an optional tone/balance controls solution, which could be physically added and turned on/off as needed...

Cheers, Paul.
 
Well, direct coupling has its own potential problems too.

True, in fact the varying conditions can be dramatic if not dealt with properly, especially when using high gm type tubes or pentodes, as you aptly pointed out. Good to get the pros/cons discussed. At some point the first set of decisions will need to be made, for example tube choice or topology, and then the rest of the design can flow around those decisions made. Seems logical to let the requirements gathering play out, then summarize and agree to them, then it will probably be more obvious which direction to go.
 
I'm a bit intrigued by a few tubes. 4P1L, C3M, and 18040s are all on my radar for this project.

I'm interested in the 4P1L as I bought some of them a couple of months ago but haven't decided exactly what to do, i.e., line stage, power amp drivers, etc. I would probably want to use them as triodes based on reading here:

http://www.bartola.co.uk/valves/dht-pre-amplifier/siberian-gen-3-4p1l-pre-amplifier/

Look through his site as he has other info on the 4P1L.

These are really nice stepped attenuators at a reasonable price. A friend in my audio club has one of the stereo units packaged in the nice case with source selection. Very nice quality.

http://www.akustyk.com/

Just throwing some ideas.

Let's go.
John
 
How about four modules (all optional except for line stage)

Phono preamp
Input selector
Tone controls
Volume control and line stage
 
As far as tubes, what's common, cheap and linear? 6CG7, 6GU7 (6V version of 12BH7), 8FQ7 if you're REALLY cheap... counted 36 of those in my tube stash last night.
 
I like this approach very much!!!
Been thinking about this for a while, and have a few ideas/thoughts on how to implement it...
Cheers, Paul.

How about four modules (all optional except for line stage)

Phono preamp
Input selector
Tone controls
Volume control and line stage
 
Some Russian tubes are also good sounding and inexpensive:

6N6P - approximate ECC99 equivalent, but not exactly;
6N23P - 6DJ8 close cousin;
6N2P - 12AX7 equivalent with 6.3v heater
6N1P - 12AU7 equivalent with 6.3v heater
6N3P - 2C51 (I think)


For a phono section, a 6F12P ( pentode/triode) in one 9pin bottle would work very well...

All tubes mentioned are currently in production, and inexpensive via eBay, or online stores, even factoring in the cost of shipping...

If we decide to use and of the 12A*7 tubes, a simple switch between 12 and 6V heater can be implemented (GordonW has recently published such schematics), so that all these tubes could be used...

Cheers, Paul.


As far as tubes, what's common, cheap and linear? 6CG7, 6GU7 (6V version of 12BH7), 8FQ7 if you're REALLY cheap... counted 36 of those in my tube stash last night.
 
How about four modules (all optional except for line stage)

Phono preamp
Input selector
Tone controls
Volume control and line stage

I have no problem with additional functionality, but I'd want to take an iterative approach, concentrating on the last item. For those compelled to do more than the line stage, I'd encourage them but I don't want the thread to go off in too many directions.

As far as tubes, what's common, cheap and linear? 6CG7, 6GU7 (6V version of 12BH7), 8FQ7 if you're REALLY cheap... counted 36 of those in my tube stash last night.

I have no qualms with picking a tube up to 20-30 a piece if it makes sense. Before picking appropriate tubes, I think we still need to figure things out.

How much gain we need in the linestage, for instance.
 
Back
Top Bottom