They tell me valve amps have to be “biased”. I always took this to be something calculated at design stage, testing the bias and fine-tuning relevant resistors after assembly, to be done again when changing valves. More importantly, only to be done by valve amp experts and magicians. I was lead to believe that this is an art only handed down from father to son, in secret. Could well be. Many ways to test, maybe only one correct way. How do you do it?
This amplifier has two small trimpots (“VR1” and “VR2” on the circuit diagram, which I could find thanks to that modern marvel called “Internet”). The amplifier also has a “CON3” block on the PCB, looking like it requires a three-pin socket to go there. I would think the factory tech has just such a fitting to do the bias. The circuit diagram says “BIAS PIN 1, 3 39mV” above this “CON3” thing. The centre pin splits through two 1 Ohm resistors to the power valve cathodes. Now, posters on the net use mA as well as mV (both 39) for this setting, some sayning you need to bias for maximum 70% of plate dissipation, hence 39 mA, etc. Of course, the owner’s manual does not mention this rather important detail (“no user servicable parts inside”). And so, I would have destroyed my valves (if not damaged already) by doing just what the amp was built for – playing it.
Easy-to-locate “CON3” .
Conveniently placed VR1 and VR2. After initial adjustment to 35.4 mV. Schematic specifies 39 mV.
I started by checking the bias. This should be correct from the factory, yes? Many internet posters tell horror stories. I concur. I clipped some alligator clip leads to the three pins, centre pin lead clipped to the earth probe of the multimeter, so all I had to do was clamp the positive probe alternatively into the leads from Pin 1 and Pin 2 as I go along, far away from the circuit itself. Less Stress.
Leads clipped to the pins. Just enough space. This is a photo mock-up, no power.
I was amazed to measure 75 mV for VR1 (Valve 5) and 48 mV for VR2 (Valve 4). The valves were being cooked in a bad way. Harsh sound indeed.
Setting the bias is an “adjust VR1, adjust VR2, adjust VR1, adjust VR2” procedure. I set both close to 35.4 mV for a start. Too cool?
(Must check HT and filament supplies as well. The circuit diagram shows the filament supply to Valve 1 to be separate from the rest, taken off the 12.6 VAC tap. Why? All three pre-amp tubes are ECC83-A? Oh, wait, I see. Pre-amp Valve 1 has DC filament voltage. Cute. Still is a noisy amp.)
Does it sound better? Maybe. How would one be able to tell at my low volume? I can say, when it is warmed up well, it does appear to be “smoother”, and I perceive it to have less bass/more treble. (But, my ears seem to get more treble-sensitive as I go along. Or is it really the amp “opening up” as it gets warm? I find this to be true with my valve Hi-Fi as well.) Is this a result of cool bias resulting in “thin” tone? But, here’s the thing, turning the speaker away from me, opening the Normal volume to 5, with the OD channel set at Gain 5 and Master Volume 8 to match, with the guitar volume at 5 – 6 (Ibanez, linear pot?), covering my ears, the system reveals some interesting and exiting dynamics, which sounds like fun, and encourages me to keep on making noise. It did not have that character, before. Fun as it may be,
1. it is beyond way too loud, and besides,
2. I cannot play well enough to exploit this newfound tone.
I really do need to explore this, if an attenuator does not work, I see myself in the market for a smaller Marshall (DSL 1, 5 or 15? Other, better options?), suitably speakered and modded. (Rob Robinette has a 1 Watt JCM 800 (and 22 Watt) circuit on his site, but after the 5F1 build, I am not feeling in the mood to go that route. Would be fun, though.)
I am still amazed by this amp’s volume. I have a 15 Watt Ibanez TSA15H, which Internet posters agree is weak in volume (it is), and the small (ha) 5F1 build, which gets to window rattling volumes, but this Marshall Haze 40 is, well, very loud when I plug the Ibanez into it.
