A bit of amp blueprinting from the past

I was going through some records on some customers amps from the past.  I came across one blueprinting session that was a little different than most as these were preamps which contained eight tubes.  There were also multiple units that were very different as the model had changed in revision levels over it's production life. 

I thought that going over some of the data would shed a little light on some of the process of blueprinting and some of the ways that tubes are very inconsistent.

Bogner Fish Preamp - 8 12AX7 tubes.

Click on "read more" for the complete story.

Things started when the tech for the player brought in one Bogner Fish Preamp for one of the folks he supports.  I was told that two of the preamps were used in a rig with two Mesa 100 watt power amps.  The two rigs operated in a very different manner, to the point where one rig was acceptable and the other was not really acceptable.

Could they be improved and could they be more alike?

In a dual amp rig I prefer the two amps to not sound identical but I do prefer when each of the two amps sounds as good as they can sound.

In this case one rig sounded mediocre and the other was even more lackluster.

A few measurements were first done to establish the difference in artwork and consider how that would enter into the equation.  Once that was complete the bigger task was started, seeing what the tubes were doing and how they impacted performance.  This was phase one.  

Below are four charts.  One for each of the two preamps; before and after the blueprinting process.  I will explain some of the points on some of the charts.  

The final tube selection was based on many factors.  Tube performance and measurements were one factor and sonic character, tone and response were other factors.  

The first step was getting an idea of just how bad things were and documenting this as a starting baseline.  

The initial measurements and traces indicated to me that things could easily be vastly improved.  In the end it turned out that I was correct.

Clicking on the below charts will expand their size.

Preamp Number 1 before blueprinting

Looking at the chart above:

The numbers of 1.2 / 1600 and 100 are design spec data for a proper 12AX7 vacuum tube.

One step above is to take the 8 triodes (16 elements) and add the plate current.  15.8mA is what is there.  19.2mA is what I would have liked to see (1.2mA x 16 elements).  

The average gain was 88.  The average gain of the other unit before blueprinting was only 83.  Unit number two was the less favored rig as a side note.  The player is a very hard rockin' player.

25600 was the transconductance I wanted to see in the above amp, or at least in the ballpark before fine tweaking.  20550 was what was there.

The Sovtek short plate WA tubes are sturdy workhorses but the least defined and articulate 12AX7 available.  They were replaced in short order for both sonic reasons and also because they were very weak.  Looking above at the individual tube transconductance where 1600 is the expectation it is clear that these tubes were not usable.

Preamp number 1 after blueprinting

In the end we had 26070 TC which was over the desired target of 25600.  A little faster a little stronger than design spec for a hard rocking player.  Gain was held in the nineties as this design cascades tube stages and at higher levels of true gain noise became an issue.  I compensated by using tubes with a faster rise time and higher transconductance.

If you want to see a relationship between transconductance and gain which many folks think are the same thing:

Preamp number 2 after blueprinting

In the end on preamp two (above) we had our gain at 92 which was very close to the 93 of the other unit.  

In order to have each unit have a distinctive voice rather than just sound like a big mono rig the tube types were altered and the transconductance was adjusted via tube selection to compliment the other unit.  The rise time of this unit was also a little slower.  In a studio environment it would be a better selection for earlier styles of rock (60s and early 70s) and blues on two of the channels.

Preamp number 2 before blueprinting

Once the amps were operating properly, the final phase which is virtually impossible to document was the selection by the player of specific types of tubes.  

The tubes were selected from a very large quantity of tubes which met design spec as closely as possible or had desired characteristics.  

Each type of tube has a unique sonic signature and this was leveraged according to the desire of the player.  Faster or slower rise time, later or earlier compression, earlier or later breakdown of plate current output for more or less clean headroom or more or less gain.  Each tube was auditioned individually comparing it's sonic signature with other 12AX7 tubes from different manufacturers.  This was complicated as there are eight tubes in these preamps and there is interaction that changes based on channel selection.

There is more information on many of these numbers in recent posts about NOS tubes and some of my recent 12AX7 tests. 

If you have questions feel free to either comment in this blog or on facebook where this link may be posted.

Thanks to the folks at Bogner who watched the process. Thanks you for your feedback and interest.

Thanks to Dan Boul and Mike Franceschini over at 65 Amps for being test players along with the tech and the owner of the rigs.