5879 SE redo

[firemedic]

Also move the 33k gridstopper to V1p2 (most important), and the 68k gridstopper to V2p1. Solder them right to the pins. Other than that it looks like you're square.

I find it hard to quantify in words the tone I find pleasing. Mine is quite chimey, and I like the very rich clean tone I've gotten here; I may really be liking the partial, half-wave distortion that is a characteristic of well-but-not-over-driven tubes (and old-school PI circuits). I do not have an o-scope so it's hard to say. I'm trying lower-mu tubes in V1 because I don't really want distortion per se.

I happen to like some mid-scoop, common to the oft-abused FMV tone stack. That said, having tried a standard 6.8k (anemic), 10k (thin) and 22k (getting warmer) midrange resistor I'm willing to bet the 33k will do the trick on this particular build. For me. For more-than-average mids I would feel confident with a 56k-82k in there too without making the midrange overwhelming. My PAB switch substitutes a 680k to ground, which IS too much.  

But we are getting ahead of ourselves, this conversation is for later, The Tweaking Phase.
 

[Willabe]

I've always wondered about this. Base pinout shows NC for pin 2, 6.

But look at page 1, bottom right, installations and applications, recommended pin 2, 6 be grounded.  Hmmm.......    


      Brad           

[John]

Phew, thanks. I did look at the data sheet, but after my debacle with the way I read the relays on the layout, I was worried!

[firemedic]

It took me a whole day to figure out how my little TOS relay was supposed to be wired. A whole day.

[John]

Well don't feel bad. I think I fried my relays; at least, they no longer make that cheerful little clicky sound. I bet Doug just loves newbies... double the parts! 

[jojokeo]

I've always wondered about this. Base pinout shows NC for pin 2, 6.

But look at page 1, bottom right, installations and applications, recommended pin 2, 6 be grounded.  Hmmm.......     


      Brad           

Good eye Brad. I'm more familiar w/ ef86 and you have to be careful w/ tube connections that say NC or IS or other stuff. I wish they would be labeled  a bit better than say IS (internal screen or shield) for example. Sometimes I think there is an internal connection that shouldn't be connected to anything from the outside? I believe I done that before by tying a grid stopper or screen resistor one time and the resulting sparks the occured upon fire-up. So it always makes me apprehensive doing so w/ out knowing for sure.

[Willabe]

I'm more familiar w/ ef86 and you have to be careful w/ tube connections that say NC or IS or other stuff. I wish they would be labeled  a bit better than say IS (internal screen or shield) for example.

Yes, it is confusing the way they are labeled from one type to another. I've seen different sheets show different things for different small signal pentodes. IIRC some have the shield connected internally to the cathode, some don't.

I seem to recall, sometimes even the same tube number/type, but from a different maker will have a different pinout --- for the shield --- as far as IC or NC.
 
 I guess you could use a DMM and check the tube pins for continuity.



         Thanks,     Brad    

 

[ncusack]

Ok so things have gotten a little more discussion over the weekend. In light of some of the comments here are a few new questions.

Not sure if you guys settled on it but do pin2 and 6 need to be grounded or not on the 5879?

joko - You mentioned the creamy tone typically comming from mids and that thats typically the "tweed" tone. That would be what im looking for with this amp and was wondering if there would be any benefit to using a "tweed" tonestack instead of the TMB?

[John]

Not sure if it "needs" to be, but I did ground mine with no ill effects.

[tubenit]

I have never grounded 2 & 6 on my 5879 amps when I built them.   

I tried it once (after I had played a build for a while) since someone suggested it and found it made zero difference & the amp was already quiet.

With respect, Tubenit

[jojokeo]

joko - You mentioned the creamy tone typically comming from mids and that thats typically the "tweed" tone. That would be what im looking for with this amp and was wondering if there would be any benefit to using a "tweed" tonestack instead of the TMB?

There's many ways to increase or maintain mids. Your 33k can be increased even more say by a switch for several settings of mids. Or you can simply put a switch btwn the 33k & ground which is called a "ground lift". You'll notice also an increase in signal strength which is nice. Or you wire a tone stack bypass w/ a switch. Or don't use a tone stack? It just depends what you want. Just remember that when lifting/bypassing or having no tone stack that at moderate to high power levels, it's very easy to have too much bass which muddies your tone. So you have to lower coupling cap and/or cathode bypass cap values depending what route you take.

[firemedic]

I finally got rid of the little fuzzy noise that made me start this thread in the first place.
Stupid silver mica tone cap.....I (errr, Tubenit) redesigned & rebuilt the entire amp because of the noise that thing made.

I assumed that since I was getting this fuzz riding the notes, with every tweek, then there must be a fairly narrow range of useful resistor & cap values (i.e. grid current, output impedances, etc.) I could use with this topology.
But I was wrong- It's wide open as far as tailoring the sound to your liking. It sounds great w/ 12AX7, 5751, and 12AY7.

Neil, is yours built yet? I know you weren't looking for more options but you should know the possibilities you have to work with....

[ncusack]

Yeah nothings been built yet i've been dabbling with another idea along the same lines. Any new information would go a long way though.

[firemedic]

Well good luck, there's only about 750 million custom schematics here to choose from.

[firemedic]

 
There is the faintest little white fuzz, it starts after the note starts to decay. It's not nearly as bad as the other noise was but it is there, at pretty much all settings, even low volume. What could THAT be?

[jojokeo]

  only wild guesses w/out more info.

[firemedic]

Another revision.
I messed with a couple values & rewired the PA again, to accommodate a 12DW7. The triodes needed to be reversed to make the AU triode do the CF job. New tone caps, & I was vewy vewy careful with the leads & ironing.
Whole different amp. With lots of high-end, so when treble is dialed back, even all the way off, there is still considerable brightness. Guitar tone control is the only real treble cut. Maybe I wired it wrong- but I like it.
It's got a ballsy sound, real in-your-face. Esp. with the PAB. The chime is still there.
I'm waiting on a backordered 50V 50W zener from mouser, when it comes I intend to drop the voltage, it's still too high for my liking....

[firemedic]

I took out the 47pF cap paralleled with the 100k V2 gridstopper. Now I have functional tone control, looks like that did it. Sounds sweet.

[firemedic]

Okay, jjasilli has turned me on to using some NFB in this thing to finally lose this fizziness once & for all. I'd really like a small global NFB loop from the OT secondary but given the unusual topology on this thing (see schem above) where would I insert it? i.e. is there anywhere on a CF I can insert it? How about the 5879 cathode, removing the cathode bypass cap?

Update:
So far my best option looks like the AA764 Vibrochamp style NFB insertion (thanks JJ), however a pentode/CF followed by a tone stack is a very different animal than a simple triode stage. I'm not smart enough to figure out the various phase shift issues presented by this topology. Y'all think it will still work?

[firemedic]

Well it did work, in fact the split cathode alone on the 5879 seemed to stabilize the circuit more than anything; the global NFB insertion didn't make a huge difference but did help. The tone is smoother now, less harsh overall & especially during the attack. Thanks jjasilli!
I still would like to lower the voltage. I doubled the B+ resistor feeding the 5879/CF but it had almost no effect on the voltage. I gave in & am using a 5881 which sucked the overall B+ down @20v- not quite enough. The Zener arrives tomorrow.

[jjasilli]

 

Where do you want to lower voltage?  If everywhere, maybe use a bucking tranny.  If only to the preamp, you can try a voltage divider; i.e. not just a series dropping resistor but a shunt resistor also.  E.g. 22K 1W series / 100K 2W shunt should knock 20% off your B+, downstream from the power tube screen supply.

[firemedic]

I need the entire B+ dropped. What's a bucking transformer?

[tubenit]

IF I were in your shoes, I'd remove the choke and add a 3.3k/2w between node A & B & 10k/2w between B & C. I think your 6V6 plate voltages are just fine personally.


Maybe you could move the choke prior to node A? I'm not smart enough to know if that would work or not but maybe Sluckey could chime in on that?

With respect, Tubenit

[Fresh_Start]

IF I were in your shoes, I'd remove the choke and add a 3.3k/2w between node A & B & 10k/2w between B & C. I think your 6V6 plate voltages are just fine personally.


Maybe you could move the choke prior to node A? I'm not smart enough to know if that would work or not but maybe Sluckey could chime in on that?

With respect, Tubenit

Fender Champ 5E1 and many other single-ended circuits have the reservoir cap followed by the choke before the power rail node feeding the plate.  Push-pull design rejects noise at plates.  Single-ended doesn't.  Also, you'd need a BIG choke to filter plate supply for two 6L6 or EL34 power tubes but not so much for a single 6V6.  The reservoir cap doesn't have to be really big.  Playing around with Duncan's PSU, the second cap value (plate/OT supply) had a much bigger affect on ripple.

It is possible to have a choke follow directly off of your rectifier (IOW no reservoir cap before the choke), but getting the specs right for that choke are much more complicated.  It's called a "choke input filter" IIRC.

Are you using an NOS 5Y3GT rectifier tube?  If not, you'll drop about 25 volts just switching to a real 5Y3.

HTH,

Chip

[jjasilli]

Bucking tranny: checkout this thread:  http://www.el34world.com/Forum/index.php?topic=11981.0  post if you need more info.

Choke:  subjective thing:  I love 'em!

[jojokeo]

Are you using an NOS 5Y3GT rectifier tube?  If not, you'll drop about 25 volts just switching to a real 5Y3.

A 5Y3 recto only supplies 125mA of current. Not enough for a pair of 6L6 or El34 pp in AB1

[tubenit]

A 5Y3 recto only supplies 125mA of current. Not enough for a pair of 6L6 or El34 pp in AB1

Silvertone 1483 has 6L6's and uses a 5Y3GT.

http://www.el34world.com/charts/Schematics/files/silvertone/silvertone1483.pdf

So in some cases, it apparently can be done. I've seen some other amps with 6L6's & 5Y3GT also but not very many & it would be easy to exceed the 5Y3GT limit just as jojokeo has suggested.

With respect, Tubenit

[jjasilli]

Maybe post a schematic with voltages of the amp as it is now.

[firemedic]

It's closed up again. Voltages-from memory:
@380V on the plates & screens,
@220v 5879 plate,
@85v 5879 screen,
@210v CF cathode
and some normal voltages on V1.

The 5879 plate & screen are higher than I want, I doubled the B+ dropping resistor- almost no change.
The CF heater-cathode voltage is too high, even for a 'U7, and I don't want to raise the CF heaters. I'm just lazy that way.

Yes it is an NOS 5Y3, Providing max drop. SE amp, not PP, so current draw is OK.
Please note there is a big 220R 10W dropping resistor after the recto, which gets HOT. It only drops about 10v and I want it out of there.

All these minor issues could be resolved w/ a 10-15% lower B+.

I like the IDEA of the bucking tranny but
1- this amp is tiny, there's no more room.
2- it drops the heaters too.
3- The Reverse-polarity 45v 50W Zener arrives today.

I really appreciate the input from wiser minds. It's good to have options (that I wouldn't have thought of on my own) to compare.
 
I too love the choke. What a difference in noise level.


 

[jjasilli]

Assuming tubenit's schematic from P1 of this thread is reasonably accurate to your build:  

*  Maybe move the choke to before the 1st, pre-filter stage.  A choke input filter will drop B+ voltage.  It will also cause voltage regulation -- less sag.  This may not be an issue in an SE amp.  Maybe worth a try.  N.B.: the Henries of the input choke must match the specs of the rectifier tube.  If an issue, I think it may be OK to isolate the tube rectifier from the input choke with a small value power resistor, say 10Ω (probably at least 3W).

*  If your 10W resistor gets too hot, go to 25W.

*  A bucking tranny is another solution, as stated before, to drop all B+.  For $9 Radio Shack gives you multi-tap tranny which can be made switchable.  If 120VAC into your PT yields about 345VAC out (loaded):  345/120 = a factor of 2.9.  With a bucking tranny: the 6V CT will drop 120VAC > 114VAC X 2.9 = 330VAC X 1.1 forward conversion factor for 5Y3 = 363VDC plates & screens.  That's about a 17V drop from what you have; which should yield about 203VDC on the 5879.  The 12V tap on the bucking tranny will double that drop.  That should get you about 185VDC on the plate of the 5879.

[firemedic]

Um, is there anything technically wrong w/ using the zener though? I mean does it suck the tone out or something? I'm planning to use it on the PT CT, but everyone seems kind of resistant to the idea.

"For $9 Radio Shack gives you multi-tap tranny which can be made switchable."
Radio Shack? Has something I can use? 

[jojokeo]

The zener is a good way to drop voltage w/ out an effect on tone. No worries there.

[Fresh_Start]

The zener is a good way to drop voltage w/ out an effect on tone. No worries there.

I used a 50 volt zener on the center tap of a Tweed Princeton (sort of) and it worked like a champ!  Sorry, I love really bad puns.

Cheers,

Chip

[jjasilli]

Princeton. . .worked like a champ! 

Nothing operationally wrong with zeners.  Except some users claim bad experiences with failure.

Re bucking trannies, read the thread I posted above + especially the site in there by Keene on "vintage power" at www.geofex.com

Zeners provide B+ voltage regulation.  I'm prejudiced against that in a guitar amp. 

[firemedic]

Got the zener in the mail today. From the mounting hardware included, it looks like it's supposed to be completely (electrically) isolated from ground? Is that right?

[jjasilli]

Please specify the specs of your zener - what specific item do you have?

[firemedic]

It's an NTE 5273AK. Cathode-to-case, or reverse, polarity.

[Fresh_Start]

Got the zener in the mail today. From the mounting hardware included, it looks like it's supposed to be completely (electrically) isolated from ground? Is that right?

No, you drill a hole through the chassis and bolt the zener's cathode down.  The chassis serves as your heat sink and that's important in this application.  Make sure that the center tap of the PT and the negative lead of the reservoir cap (very first in power rail) are the ONLY things grounded on the anode (loop) of the zener.  They will be at -45 volts with respect to the chassis and the rest of your grounds.  The highest current in your amp cycles through the PT secondary and the reservoir cap, so they should always be tied directly together.  It reduces noise IME.  Everything else in the amp should be "grounded" to the chassis as usual.  Also, clean the metal and scour it with sandpaper or something around the hole for mounting the zener on both sides of the chassis.  You want the best electrical connection possible here.

That's one expensive little critter unless you found better prices than I did!

http://www.alliedelec.com/Images/Products/Datasheets/BM/NTE_ELECTRONICS/235-0139.PDF

Cheers,

Chip

[jjasilli]

There is no PT center tap; this is an SE amp.  I don't where to put a reverse zener in an SE circuit. 

[jojokeo]

There is no PT center tap; this is an SE amp.  I don't where to put a reverse zener in an SE circuit.

You're thinking of an OT/output push-pull tranny vs a SE OT tranny. The PT/power tranny B+ secondary is the center tap for this.

[Fresh_Start]

The attached schematic may help.

Chip

[firemedic]

Hey guys, I went ahead last nite & installed it, prior to reading your posts. The reservoir cap is still tied direct to chassis & I will change that ASAP. Otherwise it's right.
It was $18 at Mouser, not too bad really. I like how the notes declare "not recommended for new designs". It gets warm but not hot.
I like where the voltages are now. I got rid of the hot B+ dropping resistor.

I don't have my voltage sheet in front of me, will post tomorrow.
6V6 biases a little cold at 33mA- plate voltage 330, 5881 at 70mA-plate voltage 275, just barely less than 100%. W/ existing bias resistors.
The 5879/CF was still high, @212v on CF cathode, 210v on 5879 plate, and 73v screen. After experimenting w/ voltage shunt from the D node, I settled on 27k to ground, which gave me 5879 plate at 165v, screen 63v, cathode 1.8v, and CF cathode at 167v. Much better.
So I'm happy with the voltages now. The fam was asleep so I won't know how it sounds till later today.

[jjasilli]

There is no PT center tap; this is an SE amp.  I don't where to put a reverse zener in an SE circuit.

You're thinking of an OT/output push-pull tranny vs a SE OT tranny. The PT/power tranny B+ secondary is the center tap for this.

Yes, whoops, sorry: late night brain dysfunction! 

[firemedic]

I went in to move the 1st filter cap ground to the zener, when I noticed/remembered that that is also the node feeding the PT. i.e. no reservoir cap.
Should I still connect that cap to the zener?
Will it throw off the power tube bias?
Or should I connect the cap AND the cathode resistor?

[jjasilli]

The reverse zener goes between the PT CT and ground (not in the B+ rail).

[firemedic]

I was referring to Fresh Start's suggestion to also connect the ground of the reservoir cap to the zener. No B+ involved except thru the cap.

[Fresh_Start]

Please look at the schematic - "G" is the connection I was referring to.

Chip

[firemedic]

I see that the reservoir cap negative side is connected to the zener. But your princeton has nothing else connected to that cap. My question is, since my OT tap is connected at my first cap, maybe I shouldn't connect the negative side of said cap, to the zener?
Because I don't really understand what the zener is doing. Just that it's doing it.

[Fresh_Start]

Voltage is relative.  Zero volts at "ground" is a convenience.  The zener provides a lower (negative) DC voltage reference for the PT ground, with the effect of lowering the PT's high voltage secondary lower relative to everything else.

HV "285" volts DC
-
-
-
Chassis - "ground" at "zero" volts DC
-
Zener anode / PT secondary center tap at -45 volts DC

Given a supply voltage applied to the PT primary, you can't do anything to change the voltage difference between the HV end and the center tap end of the PT secondary.  It's still 330 volts (a guessimate based on schematic voltages).  The zener shifts that voltage range relative to everything else in the amp.

285 - (-45) = 330

I suggest adding an isolated reservoir cap to the power rail if you can.  If you can't, you'll have to experiment (unless someone else has a better idea).  The PT center tap, the filter cap for the OT tap (supplying power tube plate), the filter cap for the power tube screens and the power tube cathode "should" all be grounded in the same place.  That minimizes the loops current needs to travel through.  However, you want the power tube cathode grounded at "zero" (chassis), not at -45 (zener anode).  Otherwise, your power tube bias will be right back where it was before. (Took me a few minutes to remember what went wrong when I first implemented the zener on the PT center tap.  It was my first amp build...)  I don't know whether you will get lowest noise with the screen grid filter cap grounded on the chassis or on the zener's anode.  I'd try the chassis, tied to the cathode, first.

I sure hope that helps more than it confuses things,

Chip  

[firemedic]

Okay, that's what I thought. 
The chassis is from a tiny, tiny little SS amp I converted to tubes & there's just no more room for another cap.
It's remarkably hum-free though, & I like the voltages, so I'll leave it as is. Thanks guys!