Forum Index Forum
Jazz Guitar Discussion
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Adjusting A Vacuum Tube Amplifier for Minimal Distortion

Post new topic   Reply to topic Forum Index -> Guitar Repair & Upgrades
View previous topic :: View next topic  
Author Message

Joined: 14 Sep 2006
Posts: 407
Location: Eureka, CA

PostPosted: Sat Feb 20, 2010 12:59 am    Post subject: Adjusting A Vacuum Tube Amplifier for Minimal Distortion Reply with quote

The Fender Bassman is a vacuum tube amplifier that is the great-grandfather of all successful modern vacuum tube amplifiers. (Some clones, like the original Marshalls, are in fact component-for-component carbon copies, it is said.)

This '66 Bassman hasn't been played regularly since the early 1980's so I evaluated the performance of the old amplifier. A quick look at the output waveform indicated that adjustment was required. After disconnecting the AC power cord and removing the amplifier chassis, the next step is to fully discharge the power supply filter capacitor(s). The filter capacitor is charged to nearly 500 volts - a lethal potential. Removing the power supply cord does NOT mean that the amplifier is safe.

Here's the means of discharging the capacitor(s). This photo shows a red jumper clipped between chassis ground and the high voltage power supply output.

Here's a photo of the amplifier (removed from cabinet). Above the amplifier, an oscilloscope display shows both the input and output waveforms at 400 Hz CW at an output power level of 40 watts RMS:

The upper waveform on the oscilloscope is the output waveform and the lower display is the input waveform. If you look carefully at the upper waveform, crossover distortion is clearly visible. Crossover distortion occurs in push-pull Class AB and Class B amplifiers. As most of us know, this is the universal configuration for tube amplifiers of more than a few watts output power capability.

The power stage consists of two tubes that are operated 180 degrees out of phase, each tube amplifies 1/2 of the input signal. For example, one tube amplifies the "upper" half of the cycle displayed on the 'scope and the other tube amplifiers the "lower" half of the cycle. The signals are combined in the output transformer and all is well. But not in this photograph.

Crossover distortion occurs as when one of the output tubes has finished (or almost finished) conducting and the other tube is starting to conduct. If the tubes aren't well-balanced, a discontinuity is observed as the waveform crosses "zero" (the mid-point of the sine wave displayed in the photo). The tubes in this Bassman haven't been replaced since I bought the amplifier in 1975, it's not surprising that there is significant crossover distortion, given the age of the tubes.

For those younger musicians, who may never have seen the inside of a Fender amplifier chassis, here's a close-up of the Bassman that reveals a very interesting "feature":

Take a close look at the board that holds most of the components. Starting at the upper right end of the board, move down and to the left about 1/3 the length of the board. Directly under the orange capacitor, the fiber board is distorted so severely that the board has about 1/4 inch gap between the board and the chassis. This is caused by prolonged exposure to high temperatures. The Bassman consumes a LOT of AC power: 95 watts with no input drive and 170 watts under full drive conditions. That generates a lot of heat! Because no components are mounted in the axis that is highly stressed by the "bend", there was no circuit damage and the amplifier still works fine.

If a pair of matched output tubes have been installed, alleviating the crossover distortion is a simple matter of balancing the plate bias current between the two tubes. In the Bassman, and most other Fender amplifiers, this isn't done very well. First, there is no way to measure plate current, both tubes are at the same potential and there is no cathode resistor. One would have to unsolder each plate lead and install current meters in series with the plate and the transformer to measure the current and adjust until both tubes were about the same.

I prefer to use a slightly different method. When I open up a tube amplifier for the first time, I always install a 1 ohm 1% resistor between cathode and ground. This provides a means to monitor the cathode current in each tube by measuring the voltage drops across each of the resistors. It's not the BEST way of making the measurement but it's one way. An additional advantage is that the small amount of resistive feedback provided by the resistors enhances the stability of the output stages and improves distortion.

There are after-market devices that do exactly the same thing as the above paragraph describes. They are very expensive.

Once the resistors have been installed in the output tubes, a voltmeter is attached to the cathode of each tube and the voltage measured between cathode and ground under near full-drive conditions. The grid bias is then adjusted so that the cathode current is about equal in both tubes. Here's a photo of the adjustment being made:

In most Fender amplifiers, this adjustment is a single potentiometer and "balances" the negative grid voltage for the two output tubes. It's not a very good way of providing the adjustment, it's an inexpensive way. The best method is to provide individual potentiometers for each output tube, which is the way that I do it when building a vacuum tube amplifier. The cost for the additional pot is less than $1 U.S.

I wasn't able to balance the two tubes with this adjustment, the potentiometer wiper had to be at an extreme position and the cathode currents for the two tubes were 102 mA and 121 mA. That's a fairly large unbalance (20%) and normally would indicate that the tubes need to be re-matched. I had only a single 6L6 in my parts inventory, a Chinese-made tube. I substituted the new tube, in turn, for each output tube but the imbalance was worse in both cases. So I decided to live with the old tubes (we old folks tend to stick together, you know).

Making a spectral measurement of the amplifier at 40 watts output power indicated a distortion level of 3%. This is quite acceptable for guitarists, even jazz guitarists at least for tube amplifiers, which are more forgiving in terms of distortion than solid-state amplifiers. It's time to remove the power supply jumper and replace the amplifier in the chassis. After doing this, I made a complete set of measurements on the amplifier which indicate that it is still performing at factory specifications, with the same tubes that were installed over 35 years ago.

It's been my universal observation, based on five decades of working on these old amplifiers, that nothing is gained by adjusting the bias conditions of the output tubes (in either "direction") from the factory recommended plate currents. Even if one LOVES distortion, the amplifier will produce a smoother, more gradually compressed characteristic when the tubes are adjusted for minimal crossover distortion and at the correct plate current conditions.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic Forum Index -> Guitar Repair & Upgrades All times are GMT
Page 1 of 1

Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

Powered by phpBB © 2001, 2005 phpBB Group