Hi all - I bought an old Luxman/Luxkit A3600 from Japan (100V) last year and have been having a ball trying to understand how it works so I can make sure the circuit is operating as it should and the amp performs at its best. The original tubes employed by the circuit were four Luxman 8045G power triodes driven by a pair of Luxman 6240G triodes and a 6AQ8 input tube. The circuit has previously had some adjustments to take 6550s/KT88 power tubes and 6CG7/6FQ7 driver tubes. I have since replaced the 20 year old KT88s for a quad of KT120s.
As it was built by a home hobbyist I've been checking over the components to see how they match the schematic (see attached). I've replaced some of the caps (being 50 years old) and bias pots with Bourns closed pots, which were a bit dirty and dicky. The amp produces music wonderfully well and I've very fond of it!
However, I've been reading about heater circuit design and note that this Luxman design has two separate 6.3VAC secondaries for each pair of output tubes (see pic below), the pair being wired in series. Each of the secondaries is rated for up to 5.5A and each KT120 heater draws just below 2A so there is about 1.5A headroom. However, I see there is no centre tap and no virtual ground using two 100ohm resistors, which I read is the done thing.
Do you think I should add them?
Note that the D306 NPN (1S1850) makes an 8V supply that leads to C310 2200uf (10V) capacitor and R308 100Kohm resistor and then on to a pin in a preamp socket (Luxman A3300 preamplifier), which isn't in use.
As it was built by a home hobbyist I've been checking over the components to see how they match the schematic (see attached). I've replaced some of the caps (being 50 years old) and bias pots with Bourns closed pots, which were a bit dirty and dicky. The amp produces music wonderfully well and I've very fond of it!
However, I've been reading about heater circuit design and note that this Luxman design has two separate 6.3VAC secondaries for each pair of output tubes (see pic below), the pair being wired in series. Each of the secondaries is rated for up to 5.5A and each KT120 heater draws just below 2A so there is about 1.5A headroom. However, I see there is no centre tap and no virtual ground using two 100ohm resistors, which I read is the done thing.
Do you think I should add them?
Note that the D306 NPN (1S1850) makes an 8V supply that leads to C310 2200uf (10V) capacitor and R308 100Kohm resistor and then on to a pin in a preamp socket (Luxman A3300 preamplifier), which isn't in use.
Hmmmm... sorry, I meant to point out that the two secondaries of interest are the ones at the bottom of the pic that feed V4 & V5 and V6 & V7 via point to point wiring. The other 6.3V winding feeds all three preamp tubes, V1-V3, which are fed via two leads connected to the preamp circuit board.
I'll start the ball rolling by saying that the heater supply must have a DC connection to audio ground.
Normally, that involves using a transformer with a centre tap which is connected to ground, or an artificial centre tap as you suggest.
I see that each of the secondary windings in question is connected to ground at one end of the winding, but am not expert enough to know if this is another "done thing"!
Hope that gives your thread a bump.
Thanks, G! Yes, terminals 12 and 16 on the power transformer (shown on the schematic) are at one end of each of the 6.3V windings and they are linked via a short wire, then there is another wire between terminal 12 and a common ground point. I note there is a centre tap for the preamp tubes' 6.3V winding, which is also connected to this ground point.
The reason I'm particularly interested is that the temps of the first power tubes connected to the heater wires run hotter (5-10C) than the other two and wondered if the unusual wiring might be the cause. I'd also like to make sure the circuit is not unduly stressing my KT120s as they are not cheap!
The reason I'm particularly interested is that the temps of the first power tubes connected to the heater wires run hotter (5-10C) than the other two and wondered if the unusual wiring might be the cause. I'd also like to make sure the circuit is not unduly stressing my KT120s as they are not cheap!
Since they're only powering the output tubes, there's nothing to be gained by changing heater wiring. Hum that gets into preamp gets amplified - for output stage, signal levels are much higher and being push-pull, what little hum might be induced may cancel out anyway.
Check heater voltages - may be a little low with the extra current of the KT120s. If it's high by more than 5%, a bucking transformer may be in order.
Check heater voltages - may be a little low with the extra current of the KT120s. If it's high by more than 5%, a bucking transformer may be in order.
Have you swapped the tubes to verify the first socket is always hotter?
If so, maybe there is some voltage drop in the wiring past the first socket.
Measure the filament voltage at each socket to see if there are any differences.
This is unlikely, but should be corrected if they do exist.
Thanks, TB! Yeah, you make good points about the difference between the preamp and output tube heater wiring. I've checked the heater voltages for the preamp tubes because they were easy! There are test points on the PCB that are easy to get to and they read between 6.3VAC and 6.5VAC, depending on the wall voltage (it varies quite a bit day to night, probably not helped by my rooftop solar PV).
R, yes, I have swapped one pair of tubes (and re-balanced them via the fixed bias pot) and this confirmed the temp difference stayed at the socket. It doesn't entirely rule out something else as the cause, though.
Anyway, it looks like I will have to measure the filament voltages if I want to rule out something fishy in the filament heating circuit so I'll go do that and report back.
R, yes, I have swapped one pair of tubes (and re-balanced them via the fixed bias pot) and this confirmed the temp difference stayed at the socket. It doesn't entirely rule out something else as the cause, though.
Anyway, it looks like I will have to measure the filament voltages if I want to rule out something fishy in the filament heating circuit so I'll go do that and report back.
Does the hot tube have more restricted ventilation? Is it closer to an obstruction? Does this happen in both channels?
If you measure the same heater voltage at all sockets, as is likely, the problem is probably ventilation.
A fan can certainly be used with a tube amplifier, and properly done, the amp will run much cooler.
If you measure the same heater voltage at all sockets, as is likely, the problem is probably ventilation.
A fan can certainly be used with a tube amplifier, and properly done, the amp will run much cooler.
You're probably right, R. Measuring each of the output tube heater pin pairs shows 6.54V on every one of them, which is nothing to worry about as the B+ (and therefore line voltage) is reading higher than nominal, at the moment (it was about 45 min ago at ~3.30pm on a sunny afternoon).
Thanks, for your advice, guys! It was enough to force me to try flipping the amp upside down and plugging it in to take measurements, which gave me much easier access to the pins I wanted easy access to. It worked so well it is going to be my go-to method in the future. 👍