1965 Gibson GA-45RVT Saturn Amp

Mark Herrmann

Physics 398 EMI

For my Physics 398 EMI project I decided to refurbish and modify a 1965 Gibson

GA45-RVT Saturn guitar amplifier. This two-channel amp has bass and treble tone

controls on each channel, has around a 45-50 watt output, two 10-inch CTS speakers,

tremolo, and reverb. The reason for choosing this amplifier was because I had one at

home that I had previously acquired for free from a relative. It was non-functional, and I

thought that it would be great to get it working and would be a great project to learn

about amplifiers. In addition, I already had the amp, so it would be less costly than

building an amp from scratch because I only needed some replacement parts rather than

all the parts.

The amp actually turned out to be in pretty good shape. Professor Steve Errede

said he had worked on similar Gibson amps and that with some modifications I could

turn this from an okay sounding amp into a great sounding amp. As it turned out, my

Gibson Saturn turned out to be a great project amp that was a good foundation for

modifications.

As previously mentioned, the amp didn’t work, so the first stage of the project

was to get the amp working. Steve suggested a number of modifications that I could

make as desired to improve its sound after I got it working the way it was. As I worked

on my amp, Steve taught me his philosophy of not making irreversible changes to the

amp except when it was clearly the best option. I also learned about how he documents

all changes and saves all the old parts in case one ever wants to return the amp to its

original state. He also taught me about working carefully and doing work neatly inside

the amp.

2
Repairs

The first thing I did was to clean the amp out a little. It wasn’t too dirty, but I

scraped some junk off the inside of the chassis and blew dust and dirt out with

compressed air. It is a good idea to have a clean amp when working on it because it

makes it easier to see what one is doing. It’s also just good to have things clean. The

first things that had to be replaced were the main high voltage capacitors, which were

dried up. I had a couple options for replacing them and decided on two 20µF 475V

multi-section (x4) can electrolytic capacitors. The reason I choose these was because

there was limited space inside the chassis to properly mount normal capacitors and I

could easily increase the capacitance with the electrolytic can capacitors, which would

boost the bottom end of the amp. To mount the capacitors I had to punch holes in the

chassis. This was one example of where I had to make a permanent change to the amp by

punching holes in the chassis. It was the best way to install the capacitors and was how

they were designed to be installed. Pictures of this are shown in the appendix. I also

replaced other high voltage capacitors in the circuit.

The first modification I made before I tested the amp. This was to install a

grounded power cord. Originally, it only had a two-prong power cord and a switch used

to determine the polarity. This is unsafe because the musician could receive a dangerous

shock if the amplifier was plugged in the wrong way. With the grounded power cord it

can only be plugged in one way, so there is no need for the polarity switch, and the third

prong of the cord serves as a ground reference.

3
 When all the main high voltage capacitors were replaced and the grounded power

cord installed, I plugged the amp in and tried it out. It worked, but not without its share

of problems. Most notably, the two 6L6 power tubes needed to be replaced. One of them

was microphonic, meaning that any vibrations to the tube caused an echoing sound to

come out the amp. The pots were also scratchy. The presence switch made a loud pop

when it was switched.

I replaced the two 6L6 power tubes with two Sovtek 5881’s that Steve brought in.

The pots were cleaned by squirting some cleaning fluid in them and rotating them to

work the cleaning fluid throughout the pot. I also discovered that one of the 6EU7 tubes

was bad and was responsible for some of the bad behavior of the amp including the

popping heard with the use of the presence switch.

The foam inside the bag for the reverb tank deteriorated and made quite a mess so

that was washed out and left to air dry. Other minor cleanup included scraping the oxide

off the pins of the tubes so they would make better electrical contact.

The connector for the footswitch to turn on and off the reverb and tremolo was

damaged when removed because apparently when it was previously worked on, excess

solder had been left on the pins and the connector was jammed in. As a result it was too

tight of a fit and didn’t come out without the plastic on the connector breaking. Steve

looked for an identical replacement but could not find one. However, he found a

connector that would work. It was smaller than the original, so to make it work we cut a

piece of metal and drilled a hole in it the proper size to mount the new connector. I

drilled out the rivets of the previous connector and drilled holes in the metal plate to line

up with the holes now in the chassis from where the rivets were. I used screws through

4
these holes to attach the metal plate over the original connector’s hole. A picture of this

is shown in the appendix.

Modifications

Besides the grounded power cord, another modification I made was to install a

10k ohm pot to be able to control the amount of current flowing through the power tubes

to prolong tube life and keep them from getting too hot. The line voltage has increased

over the years, and so nowadays older amps are pushed harder than what they were

originally designed for. I installed a 1-ohm resistor on each power tube. By doing so, the

voltage can easily be measured across this resistor and by using Ohm’s law, V=IR, with a

resistance of 1 ohm, the voltage is equal to the current. The pot can then be adjusted to

allow the desired current flow through the tubes.

The biggest modifications I made to the circuit dealt with some Twin-T circuits in

the amp. A Twin-T circuit is a filter that has a drop-off in frequency response at the

frequency given by f = 1/(2πRC). Figures 1 and 2 show a sample Twin-T circuit and

frequency response.

Figure 1 – Twin-T schematic

5
Figure 2 – Twin-T frequency response

6
 There was a Twin-T circuit in the normal channel and one in the reverb channel

that caused a large drop off in frequency response around 370Hz as well as caused weird

phase shifting. There was one that was part of both channels that caused a huge drop off

in frequency around 70Hz. These circuits make no sense in the amplifier and can easily

be considered design flaws.

To confirm that these circuits agreed with theory I hooked a function generator to

the input of the amp and observed the output on an oscilloscope. Combined with aural

observations it was very clear that there was a large drop-off in frequency response at

70Hz. A noticeable frequency response drop-off was not very noticeable at 370 Hz.

Perhaps the effect of the Twin-T circuit there was either more obscured or less drastic.

The next step was to model the part of the amp’s circuitry containing two of these

Twin-T circuits with a computer software program called Electronics Workbench. I was

able to do a frequency analysis of the circuit without modifications as shown in figure 3

below. I then modified the circuit on the computer to remove the Twin-T circuit around

70Hz. The large drop-off in frequency around 70Hz disappeared as expected. This is

shown in figure 4 below. I returned to the original circuit and removed the Twin-T

circuit around 370Hz. The large drop-off in frequency around 370Hz also disappeared as

expected. This is shown in figure 5 below. Then I removed both Twin-T circuits from

the simulation and the frequency response leveled out as shown in figure 6 below.

7
Figure 3 – circuit without modifications

8
Figure 4 – circuit with Twin-T network around 70Hz removed

9
Figure 5 – circuit with Twin-T network around 370Hz removed

10
Figure 6 – circuit with Twin-T network around 70Hz and 370Hz removed

With all of this evidence I decided to remove these Twin-T circuits from the amp.

By looking at the schematic and following components around in the chassis I was able

11
to locate these circuits. Steve marked some numbers on the capacitors to designate which

hole they came out of, as well as drew the circuits on paper in case the original circuits

were to ever want to be restored. After this documentation of the circuits was completed,

I removed them. I put in a 1M-ohm resistor in place of the 70Hz Twin-T circuit to keep

the gain from become too large. There were also a couple extra capacitors to ground that

I removed.

I also replaced four other capacitors with larger capacitance ones, 250µF, and two

other capacitors with larger capacitance ones, 25µF, as suggested by Steve to further give

the amp a solid bottom end. The appendix contains a schematic with all of the

modifications indicated.

The final modification was to solder the speaker cable to the speaker jacks to

provide a solid electrical connection. The connection with the clip that slid on had a lot

of oxide and was not very tight which could have caused a flaky electrical connection.

Conclusion

With all of these modifications the amp sounded better. The most noticeable

change to me was the presence of a bottom end. There simply was not the bass response

with the Twin-T circuits that there is now without the extra circuitry. The amp is also

quite a bit louder now. I’m happy with the way it sounds. It has a lot of reverb available

which could create a great “surf” sound. Steve was right about it being an okay amp that

could be a great amp. I learned a lot about amps by working on this project and would

like to give a big thanks to Professor Steve Errede for all of his help.

12
Appendix

Figure 7 – Amp Chassis before removing Twin-T filters and replacing capacitors

Figure 8 – Amp Chassis after removing Twin-T filters and replacing capacitors

Note in figures 7 and 8 where I installed the electrolytic can capacitors, which appear as

brown circles on the left side of the pictures. On the far right is the metal plate I installed

to hold the new footswitch connector. Note also the 10k ohm bias pot I installed on the

left side as well as the gray 1-ohm resistors on the power tubes. By comparing figures 7

13
and 8, one can see all the extra circuitry that I removed due to the Twin-T filters, as well

as the new higher valued capacitors I installed, which are black.

Figure 9 – Rear view of Gibson Saturn after completion of repairs and modifications

Note in figure 9 the electrolytic can capacitor I installed that is hanging down just to the

right of the transformer on the left side of the amp. There is another one behind the

transformer that cannot be seen in the picture.

14
Figure 10 – Front view of Gibson Saturn

15
Figure 11 – Schematic of modifications

16