[This post continues from Part 1 of the repair of a vintage Airline 9005 guitar amplifier]
After restoration, the Airline 9005 guitar amplifier sounded good – but was subject to unduly long warm-up times and intermittent shorting out due to the unique design of the biasing arrangement of the power tubes. The amp had cut out more than once during usage, and so the owner brought it back to me for another look.
Looking at the biasing design of the power tubes – it seemed to me somewhat unreliable, introducing too many points where the bias could be broken – causing the power tubes to cut out.
I proposed a modification.
The Airline 9005 Modification
Because the biasing design was likely done for cost-savings rather than any desirable sonic quality it imparted to the player, I decided to undo the bias design altogether and replace with a new design.
The first modification was to install a 6.3v filament transformer to directly power the two 12AX7 tubes. I guessed that the power transformer on the Airline 9005 was not spec’d to handle the additional current of these two tubes – so needed to spec a new transformer.
Based on 12AX7 datasheet specs, the 12AX7 powered by 6.3v consumes 300 mA. So to power two tubes I needed a 6.3v filament transformer rated to handle 600 mA. I chose the Hammond P-T166G6.
I was able to mount the transformer on the backside of the bottom chassis in such a way that it would be tucked away and not obtrusive. The primary wires went in through the side and the secondaries rather unceremoniously straddled the power transformer and went in through the top.
On the inside of the chassis, the secondaries were formed into a twisted pair and routed to the first 12AX7, then out to the top chassis, where they were connected to the second 12AX7.
Although this did add a third connector between the top and bottom chassis, it could be physically separated from the other two so as not to introduce interference.
Once this had been installed, the next step was to bias the power tubes. I decided to start with 150 ohm resistor and a 22µf capacitor.
I later dropped this value down to 120 ohms.
This new biasing design results in significantly faster warm-up times (from 60-80 seconds before to the more typical 10-20 seconds) and improves overall stability.
Additional Mods
I also noticed that the B+ voltage was about 10-15% lower than the schematic specifications. Instead of 330v I was getting 295v – and this difference manifested itself all down the line of DC supply voltages. Just out of curiosity, I put in a 5AR4 (GZ34) tube in place of the 5Y3. The voltages immediately were back on spec. This introduced some interesting “tube rolling” possibilities for this amp:
1. Safe and Long Lasting: 6L6GC and 5Y3
Using a 5Y3 rectifier tube introduced lower B+ voltages than the schematic specified. Substituting the 6L6GB with a higher-powered 6L6GC tubes would run the tubes at a colder bias of around 55-60% that would extend the life of the output tubes. The amp sounded quite good in this mode – and when the owner first tried it out he was happy with the sound.
2. Kicking It Up A Notch: Use a 5881/6L6GB as power tubes OR a 5AR4/GZ34 as the rectifier tube
Using a 5AR4/GZ34 tube to increase plate voltages OR using the specified 5881 or 6L6GB tube would run the amp at a slightly hotter bias (~65-70%). This combination sounded a bit more lively than the first.
3. Hot and Heavy: A 5881 with a 5AR4/GZ34
Running with BOTH 5881 or 6L6GB output tubes AND a 5AR4/GZ34 rectifier tube offered a hotter bias of closer to 85%. This combination sounded GREAT – especially when amp started breaking up at higher volume. The first time I tried it, it brought an instant smile to my face as the sound came alive and was just so right.
The tradeoff is that the output tube lifespan would be reduced in comparison to first two options.
Thoughts on this modification? Leave them in the comments.