Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
One of my amps is dead flat pegged at zero when the DC offset switch is tripped, but the other is staunchly at 2 volts. I had a 6as7 blow some time ago and replaced it but no change. Tried measuring all tubes and replacing the lower performing ones and balancing measurements from one bank to another in the two rows. Tried moving the 6sn7s about. Tried removing the bottom plate and measuring the large and small resistors on the 6as7 tube sockets but no dice, they all measure about the same. The trimmer pot does not make it move more than about 1/4 of a volt if that much no matter if turned to extremes. Sound wise it seems as if the other amp is doing all the work so the sound seems localized to its speaker, but if you approach the speaker for the bad amp you hear sound, just not in the same intensity as its brother. Not sure if that could be an effect of bad DC offset or if I have a different issue. Yes I tried switching the cables to see if it were the preamp or the interconnect.
Not sure where to look next. Help.
Tin-eared audiofool, large-scale-Classical music lover, and damned-amateur fotografer.
William Bruce Cameron: "...not everything that can be counted counts, and not everything that counts can be counted."
Take a look at the link below:
I tested all the power tubes with my Hickok tube tester to make sure no shorts or grid leakage are present and even matched the tubes so that one bank of power tubes has equal readings to the other bank of tubes. Will the visual inspection matter after that? I replaced three tubes that tested poorly to ones that tested well and can test them again to see if they died after being put in circuit. Not really sure why the soft sound or if it even has anything to do with the DC offset issue.
I guess I can also retest the 6sn7 tubes to make sure they are still good. Mine has two banks 6 power tubes per bank and five 6sn7 tubes in a row starting by the power plug which is where the DC offset meter is and headed across the back from left to right. Are these tubes set up to be top and bottoms of a circuit where one has a greater effect on sound? Just casting things at the wall at this point.
The next thing to check is that all the fuses are good on the front panel.
I really should ask though- when you swapped the interconnects left for right, was that at the back of the amps and what happened?
Older 12 tube but brought up to 3.1 status. I first switched the cables at the preamp side to see if it was the preamp but got the same lackluster performance from the same channel and then switched cables at the amp side to see if it was a bad cable but it was the same lesser sound from that amp side. I will try the speaker cables next to make sure I don't have a dirty connection or that the cleaning lady did not knock something loose from the Zero autoformers.
Are you OK with checking voltages as you did in the M-60?
You don't have to remove the bottom cover. Just take a power tube out of the socket from the left bank and one from the right bank. Then locate the index of the socket and count 5 pins clockwise; that is one of the grids of the power tube. What is the voltage? You can measure it with respect to the chassis.
We need the grid voltage on the left and right banks.
Just to make sure it was not the tubes, I swapped all the tubes from the poorly functioning amp to the one working well and vice versa. Still got the 2v offset reading on the left and still got a pinned to zero reading on the right. Sound wise still have the powerful right as compared to the left.
Pulled the tubes and measured the grid voltages on grid 2 to chasis and got 410 on both rows. Tried readings comparing anodes to cathodes and grids to both on standby and fully on and despite small or drifting values as the amps warm up there is little difference between them and the readings are almost the same as the functioning amp. I only tested the two end tube sockets of each amp and maybe it would be different if I tested all the sockets.
Just to be clear, I pulled all the tubes including all 6as7 and all 6sn7 before measuring. Do I need any tubes to be in place to do the measurements? Also, you say the two banks are six left and six right, do you mean that drawing a line down the center from front to back, the two rows of six are not divided into each row, but instead comprise two rows of three tubes on the left and two rows three tubes on the right? This makes a difference for me as to how I balance the power of the two banks against each other.
I did not originally measure the other tube sockets other than the end two on the right and now measured the two sockets on the left and got the same 410v.
The 410V reading is incorrect as there were no tubes.
I reinserted tubes to start measuring bias and thought again about the fuses. Visually they all looked intact but when I used the continuity tester one of the two 5A slow blow fuses on the front of the amp was not making the short and the ohmeter showed infinite resistance. Strangely when I measured grid voltage on all the tubes at pin 4 they all showed 184-5 on the left and 181-2 on the right, even without the left fuse in circuit. With the fuse out I depressed the voltage offset switch and even without that fuse it was pinned at zero. I put a new fuse in to its place and listened and it sounds good. I think that pulling and reinserting the fuses or the tubes must have cleaned some crud off of something to make the difference. Otherwise the tubes that I thought had to be balanced one row of six to the other row I now know have reinserted to be balanced left to right. Now offset is at zero and sound is balanced between the two amps. No explanation readily available but sound is back.
As an aside to Ralph, are there any different roles played by the five 6sn7 tubes that would let me know which needs to be least microphonic, which needs the best plate to plate match or if any of the tubes need to match one another. On my amp, looking from the front, the rightmost tube is next to the xlr jack and the leftmost tube is next to the offset meter.
Thanks again for the advice.
Glad you got it sorted! FWIW we no longer use those fuses on the front panel by the power switch, and disable them when we see them. They lead to confusing issues like this one :)
The 6SN7s closest to the input connection will have the most effect.
The last 6SN7 in the row should be either a -GTA or -GTB version.
Thanks! Do those fuses serve no function? Would it hurt to put a larger value fuse in to avoid burnouts?
Thanks for the info on the 6sn7's - I take it there is no particular need for tube matching or plate to plate matching except for perhaps the first tube?
Yes, we no longer use those fuses. So we bypass the fuse holder and remove them.
The input tube does seem to work better with matched sections.
I think you need to leave the input and driver stage tubes (6SN7s) in place when measuring the output stage grid voltages. "410" is not plausible for grid voltage wrt ground.
And on pin 1 and 4 of every socket was 410 compared to chasis ground. Not sure if the 6sn7 sockets should be tested or where to look next.
ask Ralph. Ralph knows his circuit better than any man alive, and he usually guesses correctly as to cause, even from long distance.
Meantime, you apparently have a volt meter. You don't mention what Atma-sphere amplifier you own, but it apparently uses 6AS7s, which means you've got a lot of 6AS7s up and running. Can you measure the plate voltage on each bank of output tubes? (There are separate supplies for each phase of the output stage. The supply for each phase is shared equally among all the output tubes on that half of the stage, so the plate voltage across one tube should be identical to that across any other, on that phase.) Having done that and if the plate voltages are equal on each phase, you could next measure the grid bias voltages; if one side is much more negative than the other, it indicates that that half of the amplifier is closer to shut-off than the other, not doing much work, which could account for that apparent low power without much distortion or any peculiar noise. The next question would be "why?"
Don't do any of this if you are not experienced working around high voltages and currents. Never touch the chassis with both hands at the same time; I keep one hand behind my back or in my pocket to remind me not to use it, when messing with a live circuit.
I have the MA-1 deluxe which is a 12 tube version of the MA-1 with a very robust power supply which I was given to understand was from the MA-2. The amp was brought up to 3.1 status and the voltage tubes were all changed to 6sn7s. I am assuming that one row of 6as7 tubes makes up one bank and when you say to measure voltage you are referring to the voltage in the tube socket of one anode to cathode in a socket on one row of 6 as compared to one anode to cathode of a socket in the second row of 6, in order to see if the PS has an imbalance between the two banks in B+ voltage?
6 tubes on the left and 6 on the right.
Is this the same as an MA-1 Deluxe? This amp has two rows of six power tubes running from right to left at the back followed by a row of five 6sn7 tubes. If split half right and half left then you are saying the left three of each row is balanced against the right three of each row?
Sort of- the tubes in each bank are in parallel, so its the bank of tubes that are to be considered, not the individuals in the bank.
The MA-1 Deluxe was built for a short time prior to the introduction of the MkII series. It had a sloping rear panel much like the MA-2.
The best way to do it would be to invert the amplifier on a safe and sturdy workbench. Then remove the bottom plate and then measure the voltages with the tubes all in place. Each 6AS7 has two discrete sections. However, before following this line of investigation, I would suggest first of all to follow Ralph's plan of attack.
This weekend I have been dealing with almost the same problem on my humble diy M60. I measured the voltages on the sockets of all tubes (tubes taken out). Everything was fine except the grid voltage on V4 (-285). I suspected there is a fault in 5K trim pot, replaced it, and the problem disappeared.
I fixed my neighbor's Novacron, which had a problem with "low power output" (much like that of the OP in this thread), and it also turned out to be due to what appeared to be an intermittent problem with a bias pot, which resulted in operating one bank of 6C33Cs in near shut-off conditions. (Bias voltage was around -80V, which will pretty much turn off a 6C33C that sees the typical plate voltage in a Novacron.) I gave the pot a few turns in both directions, then re-set the bias, and that alone seemed to cure the problem, which has not recurred. Either that or he had just erred in setting bias. He seemed to know how to do it, so I tend to think the pot was faulty, but I really cannot be sure.
Post a Followup:
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: