Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: 12AX7 on February 18, 2012, 10:02:25 am
-
Heres one of those issues that makes you scratch your head in disbelief. My amp, a 2204 ish build with no cold bias stage works fine. But today i thought i'd try the 12ax7 mullard RI's that i'd used in it before for months at a time. I tried one in V1 and i got what sounded like a stage where you forgot to add a grid to ground resistor and you hit the strings and get a staticy broken up nasty sound that drops in volume to nothing and stays there. So i figure ok, the tube must have been ruined in the drawer so i stick a second mullard RI in and same thing. Then i start thinking what i could have tweaked in the initial stages since i last used these that caused this, but the initial stages are the same aside from a 2.7k cathode on V1B in place of the 1.5k. So i put the chinese back in and all is well. then a start trying other tubes and a EH does a similar thing except instead of dropping to no volume it just kinda sounds wacky for a second then the gain drops till it's very low and clean sounding with the gain set up high. Put a JJ in, no problem, But a few other tubes in, NP. I thought the socket maybe, but it's a belton and why wouldn't it do this with all tubes if thats it? Tried deoxit in the mullards to no avail. Any ideas?
-
> like a stage where you forgot to add a grid to ground resistor
Maybe some tubes need it more than others?
Maybe there's an almost-good solder joint that makes/not-makes with different pins?
-
I thought of the solder joints too. I'll hit em all right now and post back with results.
EDIT: hit every joint and tightened the pins earlier. All i can think of at this point is a bad cap that is on the edge somehow and certain tubes set it off. At times the tube will allow somewhat normal sound that doesn't drop, and it's due to gain. In other words, if i drop the gain knob way down at some very low setting the tube will then start working relatively normal. Very bizarre. the good thing is i love the amp as is with other tubes.But this just has me bothered because it suggests there may be something wrong which may keep it from being at it's best even with the tube that works right.
-
Sorry to bring back this old thread, but i found the reason it was doing this and wanted to see what anyone has to say. I have no idea why this would cause it. Anyways, i have used 10k to 68k grid stoppers at the input in the past, and the last time i changed i went to a 68k. That must have been around the time this all started. In any case, i changed back to a 10k and now the mullard RI's work ! Any idea why this would be?
-
The particular 68k's you used could have been bad, or maybe had a broken lead.
Or, maybe that was the location of the bad solder joint or poor mechanical connection.
Which reminds me... I've got a guitar with a shoddy connection at its jack that I need to fix, which makes the exact same sound you're describing. I had wondered a couple times when I heard it if I had a bum tube, an error in my new amp build, or a cable that's giving up and needs some length cut off the end.
When the guitar did the exact same thing with a different cable and a different amp with different tubes, it became clear it was the guitar's jack.
-
The gridstopper resistance is added to the grid leak resistance for your total effective grid leak value. +10k would have an almost negligible effect but +68k may push it over the edge to exceeding the mak gridleak value for the tube. It's not the same as having no gridleak resistor, but the idea is the same if you consider no resistor to be "infinite" ohms.
I'm still kind of surprised that even a 68k gridstopper would push the tube into cutoff though, that's a bit odd...
-
I don't think it was a bad connection because i have redone the jack side of the resistor at least if not the socket side too. I'm glad it works now tho because i really love the sound of the amp with that tube in V1.
-
> 68k may push it over the edge to exceeding the mak gridleak value for the tube.
Max gridleak for healthy 12AU7 12AT7 is _1 Meg_, nowhere near 68K.
(12AX7 sheet shows no max, though we know it gets loose above a couple Megs.)
Always possible a _sick_ tube upchucks on 68K and not-so-much on 10K. But then the problem is the tube, not the resistor.
-
> 68k may push it over the edge to exceeding the mak gridleak value for the tube.
Max gridleak for healthy 12AU7 12AT7 is _1 Meg_, nowhere near 68K.
(12AX7 sheet shows no max, though we know it gets loose above a couple Megs.)
Always possible a _sick_ tube upchucks on 68K and not-so-much on 10K. But then the problem is the tube, not the resistor.
Not unless it's coincidence, as i have 2 of them and when it first happened the first thing i did was try the other one. They both did it.
-
>
Max gridleak for healthy 12AU7 12AT7 is _1 Meg_, nowhere near 68K.
Always possible a _sick_ tube upchucks on 68K and not-so-much on 10K. But then the problem is the tube, not the resistor.
I had mentioned the stopper plus the gridleak would be the effective actual grid leak resistance. But there's not much difference between 1M and 1.068M so my hypothesis was admittedly weak.
-
Well, whatever the issue it's got to be something in the tube's design because i have used no less than 6 or 7 other brands in there and all work fine.