6072a and 12au7 substitutes

Hey folks,

Curious to know what options there are to substitue 6072a and 12au7 tubes in a mic preamp.

I have been reading that a 12bh7 should drop in for the 12au7. What characteristics might be expected? Is there anything to look out for in terms of pinout or wiring?

How about the 6072a?

Are there octal tubes that can suit?

What I'm looking for is better transient response. I think this is most critical for V1 (6072a), which tends to round off the transients at normal input levels.

Thoughts?

DB

BluegrassDan said:

Hey folks,

Curious to know what options there are to substitue 6072a and 12au7 tubes in a mic preamp.

Click to expand...

What mic amp?

BluegrassDan said:

I have been reading that a 12bh7 should drop in for the 12au7. What characteristics might be expected? Is there anything to look out for in terms of pinout or wiring?

Click to expand...

Same pinout, same gain, but twice the heater current draw so make sure your power supply can handle that.

BluegrassDan said:

How about the 6072a?

Are there octal tubes that can suit?

What I'm looking for is better transient response. I think this is most critical for V1 (6072a), which tends to round off the transients at normal input levels.

Click to expand...

There are probably octals that are similar but I don't think that would get you what your looking for.
the 6072 is also known as a 12AY7. Tubes don't have "tone" or transient response.  It all depends on the circuit they are in and if it was designed correctly for that tube. The tube is just a voltage amplifier. a 6027 will sound exactly like a 12AT7 if the circuity around each one is designed properly for that tube.

That said, You can try trading out the 6027 with a 12AX7 (more gain but less drive capability) 12AT7 (less gain but more drive capability) and see what that sounds like. If the circuit was designed around the 6072 the substitutes will not behave properly but may sound better or worse to you subjectively. Both the 12AX7 and 12AT7 would be a pin for pin replacement and fine (in most cases) to experiment with. 

BluegrassDan said:

What I'm looking for is better transient response. I think this is most critical for V1 (6072a), which tends to round off the transients at normal input levels.

Thoughts?

DB

Click to expand...

What makes you think the 6072a rounds off transients? These tubes are good to VHF. It is not hard to make a tube mic pre flat out to 150KHz.  A tube mic pre usually starts with a transformer and I would put that as the number one suspect for rounding transients.

Cheers

Ian

I'm using a Jensen JT115k. Would something like the JT110k (1:8 ratio) make the type of difference I'm talking about?

BluegrassDan said:

I'm using a Jensen JT115k. Would something like the JT110k (1:8 ratio) make the type of difference I'm talking about?

Click to expand...

With correct source and load impedances the JT115K will go up to 40KHz. Should be plenty for decent transient response.

Moving to a 1:8 may improve this a little but I would be surprised if you could hear it.

Have you measured the frequency response?

Perhaps the fault is in the circuit deign. Can you post a schematic?

Cheers

Ian

I bet he is talking about the Hampton tube mic preamp.  8)

There might be too much gain happening somewhere and your clipping the circuit. Make sure your plate and cathode resistors are the correct values. If you remove C1 it will lower the gain of the first stage (if it is the Hamptone ciruit).  Have you ran a tone through it? usually you can hear a 1 k tone distort pretty easily.  You must know the correct level  to inject. If I'm testing a mic pre with a line level out from Pro Tools, I use a -25db U pad in an XLR barrel between Pro Tools and the mic in of the pre. In addition turn down the outgoing track about 15 or 20db. Then I'm close to a mic level and impedance going in. If you have a scope even better, you can look at your wave form after the first stage at the top of the volume pot and then at the output to see where you clipping, if thats the problem.

Below is a U pad diagram. You can just use two 750 ohms for R1 and a 150 ohm for R2 to get in the ball park

Yes, it's the Hampton schematic.

Don't get me wrong. There's no clipping "problem." Just trying to experiment and tweak things slightly.

Square waves show pretty good response, especially after adding a zobel to the output. Frequency response is nice and flat. Sounds big and airy like a good tube preamp.

Of course, it has less clipping when padding down the input. But I'm just trying to find a way to get a bit more snap out of the transients.

On a side note, I've made a few minor tweaks to the original schematic. Removed C8, added 330 ohm grid stoppers, included a low shelf filter, etc.

C8 in the Hamptone circuit will definitely slow up transients. Adding grid stoppers will only make this worse. The 6072a ia a low gm tube so it does not need grid stoppers. Take them out and remove C8 and see if that improves things.

Cheers

Ian

Ian, you think it will be okay without grid stoppers? No potential oscillation?

BluegrassDan said:

Ian, you think it will be okay without grid stoppers? No potential oscillation?

Click to expand...

None at all unless the layout is really bad. You only need grid stoppers with high gm tubes like the ECC88 and output pentodes.

Cheers

Ian

Ian, you are so very generous with your help. I appreciate your advice!

Are you using an Edcor transformer for the output? Try no output transformer. That might make it sound more "snappy". It might take away some of the character but it might add a bit of clarity. Worth a try. 

Here is a statement of the designer about C8 and ways to upgrade the circuit:

"Add a totem pole follower after the first stage, before the volume control. This buffers the first stage, allowing the pot to be reduced from a 100 K to 5-10 K. Pots have reactance, the bigger the value of pot, the worse the effect on the audio signal. They can cause phase shift and dispersion if the value is too big (making the sound a bit "mushy"). In this design the first gain stage has to drive the pot without a buffer. Because of the higher output impedance of the first stage (compared to the totem pole), the 10 pF feedback capacitor is required on the first stage to keep it from oscillating at higher gain settings. If the first stage is buffered like the second, the cap can be removed for a slight sonic improvement in the top end."

I would use a grid stopper for V1 and try to get rid of C8. *edit: what Dan exactly did!

Strange, about a million Fender and Marshall amps have run a similar front end topology (with higher gm tubes) without any plate-to-grid feedback (e.g. C8).  I don't see how C8 would stop oscillation unless lead dress was pretty atrocious.

Also, am I reading that schematic incorrectly, but it looks like the second grid (V1 1/2) has no DC bias?

I must have 30+ types of tube preamps and not a one has a grid stopper. 

BluegrassDan said:

Ian, you think it will be okay without grid stoppers? No potential oscillation?

Click to expand...

Matador said:

Also, am I reading that schematic incorrectly, but it looks like the second grid (V1 1/2) has no DC bias?

Click to expand...

Yep. I think it is good practice to put a 1M resistor to ground between the pot and the grid, too. I will do it in my build.

rock soderstrom said:

Yep. I think it is good practice to put a 1M resistor to ground between the pot and the grid, too. I will do it in my build.

Click to expand...

Can someone explain the reasoning/effect of adding this resistor?

BTW, removing C8 was not merely a "subtle" improvement to the high frequencies. It was like removing a heavy wool blanket. Huge improvement!

BluegrassDan said:

Can someone explain the reasoning/effect of adding this resistor?

Click to expand...

It is important to ensure there is always a dc path the 0V from the grid. This design uses the pot for this purpose but the wiper at some point may come adrift of the track and this connection is lost at which point the bias of that  tube is undefined. It is therefore good engineering practice to add a 1Meg resistor from the grid to )V to endure there is always a connection no matter what the pot does.

Cheers

Ian