What to do with unused triode ... suggestions?

[shooter]

fwiw;
when I tweak my B+ rail, I ballpark the voltages based on expected signal, so V1 doesn't need a lot of plate vdc, the PI does.  I also like a tight NO sag rail, so my 1st filter cap(S) tend to be ~~100uF, I add to each node when I want to get ripple vac down even more.  I build xSE so anything more than ~~2vac at the 1st node will probably cause some hum.  My V1 node I like to try and get <10mVac idling.
On PP amps the PA gives you a natural cancellation of ripple because the tubes are outta phase

[92Volts]

Duncan's PSUD2 can be fun to mess around with: http://www.duncanamps.com/psud2/

Multiple stages of filtering isn't just for stage separation, it's the best way to reduce noise within your cap space/budget. Before using that software I didn't really understand what an RC filter does compared to connecting something straight after the rectifier... even if the rectifier is directly followed by huge cap you get an obnoxious "sawtooth" ripple which is smoothed to a sine wave with even one added resistor/cap. Yes, PP output stages can sometimes be used unfiltered.

fwiw;
when I tweak my B+ rail, I ballpark the voltages based on expected signal, so V1 doesn't need a lot of plate vdc, the PI does.  I also like a tight NO sag rail, so my 1st filter cap(S) tend to be ~~100uF, I add to each node when I want to get ripple vac down even more.  I build xSE so anything more than ~~2vac at the 1st node will probably cause some hum.  My V1 node I like to try and get <10mVac idling.
On PP amps the PA gives you a natural cancellation of ripple because the tubes are outta phase

Yes, PP amps cancel ripple fed into the output stage but as you know, ripple in the preamp will be added to your guitar signal and you'll hear it no matter what output type is used! So you can get away with not filtering before the OPT center tap but doing some filtering there (even as low as 50ohms/100uf if you don't want to lose much voltage) is a good opportunity to begin to lower the ripple which needs to be low by the time it hits the preamp.

You don't need to go this big, but caps can be pretty affordable these days if you're willing to wirewrap/solder "snap in" PCB caps. Here's 470uf, 450v for under $7: https://www.mouser.com/ProductDetail/United-Chemi-Con/ESMQ451VSN471MR45S?qs=sGAEpiMZZMvwFf0viD3Y3Wd4FpOoUsJZFxfSsu%252bAVZQ%3d

So despite everything I said about designing your stages correctly, overkill is also an option 

[sluckey]

Please don't put a big ass cap on that 5Y3! You have sufficient filtering already for that amp. You only need to shuffle nodes a bit.

Your latest schematic only shows 3 nodes, but your hand drawn pic shows 4 nodes. I can see a JJ 40-20-20-20 can on the chassis and that's all you need. Just wire it like this attached schematic and fix your schematic or call this one the final.

[PRR]

> The very popular and successful Fender AB763 amps supply SIX stages with the same B+ node! I always thought that was a bad idea and would never design anything like that. But it served Leo quite well.

me> Conservative design says no more than two stages on the same B+ filter. (This can be bent, but not in ignorance.)

We agree. Using "free" time to build ONE amplifier, simple rules like "2 stages per B+ node" keep us out of trouble. When you know you are going to sell tens of thousands, then you get a sharper pencil. Interstage losses "might" exceed forward gain and B+ node sneak-back, and allow you to save thousands of pennies.

BTW, I did know we had 1st and 2nd preamp AND reverb recovery on the new B+ node I proposed in reply #30. 3 stages. Bad? It's a tricky point, I think it is allowable, AND it won't motoboat until Reverb is turned-up so can be diagnosed with one finger.

I did NOT know two *sequential* stages had the SAME cathode network. That was NOT on the "more or less updated schematic". I hate getting lost in Brooklyn because I'm looking at a map of the Bronx.

[shooter]

Please don't put a big ass cap on that 5Y3!

 
fwiw, I never use tube rectifiers, well, once, the rest SS
 

for those lurking n learnin, you have available here the top shelf overlords, I did well for 30yrs fixin very complex systems, my "ace in the hole" were guys like Sluckey, PRR, HBP, and more. they are the "notch above" we aim for

[sluckey]

I did NOT know two *sequential* stages had the SAME cathode network. That was NOT on the "more or less updated schematic". I hate getting lost in Brooklyn because I'm looking at a map of the Bronx.

  Took a long time for that little nugget to surface! Coulda saved a bunch of time and head scratching if that had been on the schematic. Coulda saved a bunch more if there had been 4 power supply nodes on the schematic too.

[Diverted]

Please don't put a big ass cap on that 5Y3! You have sufficient filtering already for that amp. You only need to shuffle nodes a bit.

Your latest schematic only shows 3 nodes, but your hand drawn pic shows 4 nodes. I can see a JJ 40-20-20-20 can on the chassis and that's all you need. Just wire it like this attached schematic and fix your schematic or call this one the final.

Thanks. Yes at this point it's changed a ton. I'm trying to update it and will post it when I do. Thanks!

[Diverted]

Thanks for the suggestions ... I've seen and built a bunch of simple amps using 40uf with a 5Y3 but I am no expert. Though I haven't killed any 5Y3s yet, I don't want it to happen ever and am always willing to take advice. I've got a bunch of 33uf/500v Nichicon caps and there's plenty of room to put one of those in the first stage, and take that 40 out of there. Thanks for the guidance.

[sluckey]

I don't consider a 40µF to be a big ass cap for a 5Y3. Some others were talking about 100µF and even 470µF. Now those are big ass caps for a 5Y3!

If you are concerned about the 40µF being connected to the 5Y3, then just put one of the 20s on the 5Y3 and move the 40 to node D. No need for additional caps.

[Diverted]

Thanks Sluckey. Since reading the post this morning I have obsessively read a million varying opinions on how much is too much in the first filter. I'll take your word for it.

One thing I may do anyway is just use a 5U4 in place of the 5Y3. The overall B+ voltage in the amp is a bit lower than in the past few versions of this build (different power transformer in this one than in the others) so I think there's room for more voltage as long as I adjust the bias. The 5V winding is rated for 3A so it shouldn't be a problem and unless I'm wrong the 5U4 can handle a larger first filter.

Again, thanks for your expertise. I am learning a ton on this build.

[sluckey]

GZ34 would be my choice of rectifiers. They have an indirectly heated cathode. That means it takes about 20 seconds for the B+ to come up after you turn on the power. 5U4 comes up in about 5 seconds. I like the slow warm-up.

[Diverted]

Thanks, I actually have some nice NOS old Mullards. I tried one in the circuit but it just pushed the B+ too high (around 445 B+ or so). With the 5U4 I'm getting around 405-410 or so. I would probably use one but I don't have any zener diodes to bring the B+ down (plus I heard they're noisy. Any opinions on that actually?)
What I do have are some CL-80 inrush limiters. How about putting one or two of those on the AC line in conjunction with the 5U4?

Thanks!

[tubeswell]

I don't consider a 40µF to be a big ass cap for a 5Y3.

Same.


Leo ran '32uF' in the stock 5F2A.


I used 40uF with an old 5Y3G in my first 5F2A - thing was still going strong years later last I heard.

[Diverted]

I have a nice stash of old stock RCA 5Y3GTs. No modern stuff, same goes for power and noval tubes.

[tubeswell]

The published ratings on tube datasheets are typically conservative IMHO, because the manufacturers wanted to ensure that people would operate tubes within parameters that would help ensure that manufacturers weren't inundated with warranty obligations.


In reality, tubes are pretty tough and can take a fair bit of 'abuse'. I say this with the caveat that (of course), the harder you run a tube, the shorter it will last, generally speaking. But running it a 2 x the manufacturers recommended reservoir capacitance rating shouldn't do too much harm to the rectifier tube. Of course, you can enhance the life of the tube by making sure the envelope is well-ventilated and not within too-confined a space. The 'work' done by the rectifier tube's plates - to pull electrons out of the reservoir cap (and other filter caps') positive poles - increases as the capacitance is increased, because your tube is having to 'evacuate' more electrons (per-unit-of-time) from the positive poles of the filter caps. So all-in-all it will run hotter, and therefore the more thought you put into planning to allow for physical space around the envelope to help with the heat dissipation, the better.

[Diverted]

Thanks for the food for thought.
I just finished building the cabinet yesterday (head only) and tried to cut the back panel as open as possible to keep the air flowing. Don't think I can get more open space in there without making it too flimsy. We'll see!


One other question:
I have since adjusted the power supply caps so the first cap's 20, then 20, 20 and finally the 40 which now feeds the first two gains and reverb recovery.
I previously had had some issues with a slight ticking of the tremolo that occurred when I dimed the volume and put the negative feedback switch into the stock Fender 5E3 position (with 22uf bypass cap in circuit). The tick would occur with the rate of the tremolo, and not constantly; only when a note was hit. Before fixing the shared cathode issue on the first gain stages and shuffling the filter caps I was able to get rid of the ticking for the most part by running a 1m resistor into the wiper of the intensity pot.
Now with the new filtering lineup, the ticking is back, but a bit stronger. I don't want to increase the 1m resistor any more because it attenuates the tremolo effect. Any other suggestions for getting rid of the tick?

[sluckey]

I was able to get rid of the ticking for the most part by running a 1m resistor into the wiper of the intensity pot.

That resistor is supposed to be on the top of the pot, not the wiper. Wire it like the attached. Experiment with the cap across the pot.

[Diverted]

Ahha!

By the way, the tremolo is amazing! It's got more of a pronounced 'thump' than the Fender tremolos I've heard. And so full and thick. Really, really nice. And from half a tube!! Amazing, great stuff. Thanks Sluckey.

[tubenit]

 

the tremolo is amazing!

I love stories with happy endings!   

Sluckey,  I continue to have a great deal of admiration and gratitude for your remarkably generous help to this forum and it's members.  You're a stellar guy!  THANK you so much for your contribution and helping those of us (myself in particular) who are lost in the fog in figuring this stuff out.

It's been fun watching this thread evolve into another Sluckey assisted success story!  Bravo!

With respect, Tubenit

[Diverted]

I continue to have a great deal of admiration and gratitude for your remarkably generous help to this forum and it's members.  You're a stellar guy! 

I second that!

[shooter]

It's been fun watching this thread evolve

The student is a WILLING participant, that always helps 

[Diverted]

OK, so I have one more final, unrelated question (for now lol).
So, the output transformer has 4, 8 and 16 ohm taps. I have switchable negative feedback that runs into the speaker jack. It's 47K which is recommended by Rob Robinette for an 8 ohm load; should I just leave it wired to the jack and make no allowances for when switching to a 4 or 8 ohm load?

[sluckey]

Hardwire the NFB to the 8Ω tap.

[Diverted]

Thanks Sluckey.
Thought I'd post a few photos of the amp which is more or less done. It's for a giant Big Lebowski fan (thus the "Achiever" and "Abide" on the guitar jack). Evil Twin (I'm a twin) is my hobby name/Facebook handle.
 I moved the LED off the front panel and put it at the rear (not shown yet in pic) because it was just too bright and distracting up front. The blue glow emanating from the back is cool though. I put the fast/slow switch up front where I'd originally put the LED.
Had some friends over and we played the amp for about four hours last night with no issues. Very quiet, sounds excellent. Originally the tone was not too bright but I added a 500pf cap across the 680K resistor running out of the preamp. That gave a much better range of high end on the tone control. Amp sounds killer with really nice bass response. Great reverb and the tremolo is trimmed up well now. No more ticks.

One question: After a few hours the power transformer was HOT. I mean, hotter than any amp I've played/built etc. I know it was on for many hours on a muggy, warm night (we were outside) but man. Worried me a little bit. I ended up putting a small house fan behind the amp to keep some air moving over it late in the night. It's got me wondering about whether I should add a small AC or DC-powered cooling fan. What do you all think about putting in a small fan? A full mains AC-powered fan seems like it would be easier to deal with, wouldn't have to rectify AC to make it work ... I do have an unused 50V bias winding on the power transformer that I could use if I could figure out how to rectify the power to the correct DC.
THoughts?

Again, thanks everyone for the help!

[sluckey]

Originally the tone was not too bright but I added a 500pf cap across the 680K resistor running out of the preamp.

First time noticing that 680K resistor. Where did you get that idea? Why? You won't find a resistor there on the original 5E3.

After a few hours the power transformer was HOT. I mean, hotter than any amp I've played/built etc.

Cathode biased amps are supposed to run hot. They are biased to run full throttle even when idling. You can cool it down some by using a 500Ω/10W cathode resistor on the output tubes. I've never felt the need for a fan on a 15 watt amp.

If you must use a fan then 120VAC is the easier circuit, but they may be too noisy for you. 12VDC computer fans are much quieter and cheap. Get one of the Geek Squad guys let you go dumpster diving behind Best Buy. A small bridge rectifier and 470µF cap connected to the 6.3VAC filament string will run the 12VDC fan at reduced voltage. This will be really quiet and push enough air.

I do have an unused 50V bias winding on the power transformer

Tell me everything you know about this winding. I suspect that rather than a real separate winding you simply have a tap on the HT winding that's not rated with enough current capacity to run a fan. If so, don't mess with it.

[Diverted]

Yes, sorry, it’s just a tap.
Cathode resistor is currently 415 ohms/10w and dissipation is around 90 percent. Thanks!

[shooter]

we were outside

how long was the extension cord
I've found some cheap 100'rs heat up my SE amp(40W) but the tranny has 100mA of "unused space" so she just runs warmer 

[Diverted]

Not too long, maybe 15 feet. We were out on the back deck👍

[Diverted]

About the 680k...
When I built the first version of this a few months ago I didn’t account for the gain drop from sticking the reverb circuit in. So another gain stage wdnt in (the one I infamously shared cathode resistor with first stage) but that made it too grainy... amp was breaking up around 2 on the dial. Putting the 680k resistor in after the coupling cap brought it down just about right.

About the heat:
Most of the stuff I’ve done is cathode biased but it just seems like this one runs hotter.. I usually use ClasscTone 40118016 (330-0-330@120ma) for this amp but the transformer in this amp was given to me so I put it in. Don’t know much about it but the hv secondary has identical ratings to the CT.
https://www.amplifiedparts.com/products/transformer-fender-replacement-power-330-0-330-v-120-ma

[Diverted]

Hi all,

Just two quick follow-ups on the tremolo circuit Sluckey suggested and that I ended up putting in the amp. Sounds amazing!

1. The foot switch works as intended and kills the tremolo. The only minor issue I'm having is that this does not darken the LED. It stays on even when switch turned off.
So would there be any issue with killing it a different way, using a DPDT switch? One half would kill the voltage to the plate; the other would switch the junction of the two grid leak resistors between ground and the wiper of the intensity pot. Seems like this would completely remove the tremolo from the circuit. Any thoughts?

2. Second, minor question: The intensity pot doesn't really do anything until I hit about six on the dial. Below that, there is no pulsing but above it, the trem effect kicks in. Maybe run a 1m across the pot to turn it into a 500K as opposed to the 1m I'm using?

Thanks!

[sluckey]

1. The foot switch works as intended and kills the tremolo. The only minor issue I'm having is that this does not darken the LED. It stays on even when switch turned off.

Does that really bother you? Put the foot switch between the LED cathode and ground. The LED will be dark when the switch is open and it will pulse when the switch is closed. This also means the footswitch must be attached for the trem to work. Unless you use a switched jack to provide the ground to enable the LED if you unplug the footswitch.

2. Second, minor question: The intensity pot doesn't really do anything until I hit about six on the dial. Below that, there is no pulsing but above it, the trem effect kicks in. Maybe run a 1m across the pot to turn it into a 500K as opposed to the 1m I'm using?

Sounds like you may be using a pot with a log taper? If so, change to a liner pot. Any better?

[Diverted]

D'oh, yes I'm using an audio taper pot; don't have a linear pot at 1M. I'll live with it for now :)

The LED is awesome and very cool, but it does get a bit bright so I'd like to have it off when not being used. No big deal either way, I can live with it.

Thanks!

[sluckey]

Just move the footswitch as I suggested and the LED will be off when the trem is off. Can't say for sure but the trem may be slow to start up. Easy to do and undo if you don't like it.

[Diverted]

Thanks Sluckey. Such a great tremolo, I appreciate the help.