New build up and running.

[nandrewjackson]

Hello. New build up and running here. Thanks to so many people on this forum for help over the years.


It's a push pull 2-6V6 , cathode biased output.


Inspired by my new love for the 6V6, and the hoffman/sluckey 6v6 plexi project. 


Diode rectified.
Long tail phase inverter.
Jcm800/2204 preamp with 1 tube reverb added.


Two ten inch speakers.


The combo cab is an old solid state marshall one twelve box I got from a friend here local, who is also a member here on the forum.  I cut out the old particle board baffle and used 1/2"  maple hardwood plywood from the local Lowe's. 


The chassis is synapticamps.


I chose eminence speakers.  8 ohm 10" red fang, and 8 ohm 10" lil buddy. I think it's a crapshoot on speakers when I haven't ever heard them before,  but these 2 are a GREAT MATCH. 


I posted a week or so ago about being worried about the high plate voltages yadda yadda yadda , well, when I started that thread, I was confused about exactly what transformer I ordered.  They were out of the one I wanted,  so I hastily chose a 2nd option,  and then I hit pay with some extra stuff in "the cart" and the bill was a good $50 higher than expected.  Moving on. . . . .


This preamp sticks with mostly "marshall" components in the preamp, a few changes here and there,  mostly due to what I had on hand. 
The bright cap on the gain knob is 100pF instead of the larger 470pF or 1000pF due to personal taste.


I tuned the NFB resistor to 47K, 100K was ice pick territory and 22K was too dark.


I started out the power tubes shared cathode resistor at 330 R, fired up the amp, recorded the voltages, calculated it out to 200% plate dissipation.  Tried a 470R, that got me 105%. Third time's a charm, a 1K landed it right at 70%.


I'm posting the schematic  and a few pics.

[sluckey]

 

[nandrewjackson]

Thanks to tubenit for the "all time great" thread about one tube reverb. 

[turtle441]

Very nice.

I'm curious (still learning the finer points of design at the moment), with a PT that had a center-tapped 6V secondary for the heaters, why tape off the center tap and create the virtual center tap with the resistors?

[mresistor]

Very nice.

I'm curious (still learning the finer points of design at the moment), with a PT that had a center-tapped 6V secondary for the heaters, why tape off the center tap and create the virtual center tap with the resistors?

It seems to me he could have used one or the other.   And looking at Merlins page on heaters I noticed this 


"Transformer Centre Tap -The traditional way to reduce hum is to use a transformer with a centre tap, and connect it to ground. A refinement is to ground the centre tap through a small flame-proof resistor (anything up to 100 ohms). This will act as a fuse if there is a short between the anode and heater pins of the power valves, thereby reducing collateral damage."


Guess I forgot about the "refinement" he mentions, and it is a good idea..  I have experienced an anode to heater short.. and it is pretty serious. 

[pdf64]

I think that the schematic shows the 100k and 82k LTP phase splitter plate resistors swapped around.
The 68k grid stopper on V2 shouldn't be necessary?
That stages' overdrive character may be smoother if the cathode isn't bypassed.
The 1M grid leak should be on the input jack side of the input stage grid stopper, rather than the grid side.
The input grid stopper is typically the noisiest resistor in an amp; its noise is proportional to its value, so changing it to a 33k resistor (as per the effective value for a typical #1 input) may be a beneficial mod.

It seems to me he could have used one or the other.   And looking at Merlins page on heaters I noticed this 


"Transformer Centre Tap -The traditional way to reduce hum is to use a transformer with a centre tap, and connect it to ground. A refinement is to ground the centre tap through a small flame-proof resistor (anything up to 100 ohms). This will act as a fuse if there is a short between the anode and heater pins of the power valves, thereby reducing collateral damage."


Guess I forgot about the "refinement" he mentions, and it is a good idea..  I have experienced an anode to heater short.. and it is pretty serious. 

Yikes, I disagree with Merlin on this 
In the event of a (power) tube failure that shorts HT to the heater, my view is that it's preferable to avoid having the short pull the heater circuit up to HT voltage. As that may damage the heater cathode insulation of every tube in the amp.
If the 0V heater reference holds then the primary or HT fuse should blow in short order.
OK, it's possible that the PT may fail before either fuse blows, but that's rare in practice, and would probably be due to user error (eg oversized fuse fitted).
I acknowledge that it's a judgement call as to whether the tubes or the PT should be sacrificed if the worst happened, but it seems reasonable to expect that the fusing should work as intended and save both (clearly the shorted tube would need to be replaced).

[VMS]

Congrats on new build!!


As this is cathode biased amp, you could bias it to 100%. 70% is commonly used for fixed-bias amps.

[pdf64]

Yes, cathode bias tends to be safe to idle closer to / at the plate’s limit.
Typically, unless the HT voltage lower / load impedance higher, than is normal for guitar amps, trying to cool down cathode bias like this may tend to result in a significant degree of bias shift at high power levels, causing crossover distortion and a thin, reedy tone. Even though it’s fine at lower power levels.

[shooter]

recorded the voltages

not on the schematic you posted 

play with the bias, each time you do, play the amp.  work your way from your 70% back to the 105%.  9/10 I like ~~~90%.  If you monitor the cathode when you're jamming, you can figure out the bias "swing"

[vampwizzard]

heck yeah! congrats!

[PRR]

The input grid stopper is typically the noisiest resistor in an amp; its noise is proportional to its value

Hiss voltage is proportional to square-root of resistance.

34k makes much sense. It is a couple uV of hiss. Most 12AX7 will give a couple uV of hiss. Guitar noise resistance is complicated but is likely to be 10k-100k for a fair spread where the ear is most sensitive. So making the 34k smaller is unlikely to reduce hiss, while letting it be larger (68k) may be audibly more hiss.

That assumes you plug the guitar to the amplifier. (Does anybody still do that??)

And yes, at later stages the hiss matters less because gain-between.

[nandrewjackson]

Turtle:  I do the dual 100 R artificial center tap and tape off the PT heater center tap because when I first started building that's how I learned to build. Several sources say it possibly saves the PT in case of tube failure.  That's good enough reason for me.   


Pdf64:  I drew the input resistors wrong. Oops. They are not wired like that. The 1M is on the input jack terminals. There is a signal wire from jack to a terminal strip, then the 68K from terminal strip to V1 pin 2. 


I do indeed have the 100K and 82K phase inverter plates switched from conventional locations.  Not on purpose,  It just turned out that way. I usually make them both 100K, but my last build was a 1 ch AB763 style with RVB, no trem, and I used the 100K/82K values. I'd better open that one up and check them, too. I guess I'll switch them around next time I pull the chassis.  The chassis will be pulled and reinstalled several more times for faceplates, and cabinet finishing, I'll add those resistors to the list.


Pdf, VKS, Shooter, this is my first push pull cathode bias personal build, I was not aware that 100% was the norm, thanks for the heads up.  I worked at a small company building ceriatone-style vox clones about 6 years ago, they were all push pull 2-EL84 and 4-EL84 cathode bias, but I was building from their playbook, not doing any calculating numbers.


 

[tubenit]

Although a very well drawn schematic,  I found some of the numbers fuzzy and had difficulty reading them.  These are marked with orange/yellow.  Can you confirm these as correct, please?

There were some other values that are unusual or atypical of what one might normally find.  Not saying any of them are
incorrect  (except wondering about the 6V6's cathode resistor value which is normally 250 ohm range-ish).  These are marked with Red/green.  Can you confirm these as correct, please?

NOTE:  I changed values here and there in ways that aren't significant IF someone was wanting to build your amp.  Such as 470k instead of 475k.  I don't feel these need correcting.

I appreciate your sharing the success of your build!  Thanks!  Can you confirm these schematic values in question, please?

I attached an editable SCH version if you'd like to use it  ………. OR you can correct the values using PAINT if you have windows.

With respect, Tubenit

[nandrewjackson]

Nice schematic,  tubenit 


I have a nasty habit of sometimes writing 1's with serif, and sometimes just a vertical line. I need to standardize that.  I have to take a picture far away and crop it out to get photo file size down to upload here on the forum, making fuzzy appearance.  I've only got a smartphone right now, but I look forward to possibly using the.sch file in the future.



Orange/yellow values:


First 12ax7 cathode combo is 1K//1uF
Gain pot bright cap is 100pF
V2 pin 3,  1k//10uF


V3 (reverb driver) input cap is 0.003uF
V3 pin 6 coupling cap is 0.1uf
V4 pin 2 cap is 0.1uF
The 2 resistors to ground after the phase inverter are 470K each.
The shared cathode resistor for the 2-6V6s is indeed 1K. I started with a 330R, plugged those numbers in bias calculator,  got 200% dissipation. 


I'll get the voltages recorded and posted here soon.
Maybe some gut shot pics too.


Would a pic hosting site work better? I used to use photobucket,  but they've gone the way of the Dodo.

[nandrewjackson]

* I've read the forum how-to about pics, but a lot of that seems to need a full computer, i.e. right-click this and that, etc.


I'm using a smartphone only.


I did a quick google of "pic hosting "


Imgur?
ImgBB?
Instagram?


Any opinions on what would be best here?
I've posted google drive links here before to no great fanfare. 

[tubenit]

The shared cathode resistor for the 2-6V6s is indeed 1K. I started with a 330R, plugged those numbers in bias calculator,  got 200% dissipation. 

Try this …………….     Take the voltage of the shared cathodes on the power tubes divide that by the ohms of the cathode resistor and then divide that by 2.   What do you have then?

[For example:  voltage on cathode is 29v divided by a 270 ohm resistor = 107 ma divided by two tubes = 53ma per tube  (in this case 5881's).] 

I've never seen anyone use a 1k on a pair of cathode biased 6V6's that I can remember?  I think this is worth double checking on?   You may have it right?

With respect, Tubenit

[shooter]

calculating numbers.

that's fun, it's the formulas that suck 

I'm with Tubenit, I have a PP self biased ~320plate, 250 ohm common cathode R setting at 20.5vdc running 82mA or ~~13W per tube

[nandrewjackson]

Here's a img.bb link to updated schematic with running voltages.


https://ibb.co/tcNMsN5


6v6 pin 3 471 V
6V6 pin 4 472 V
6v6 across shared cathode R 40 V
6v6 pins 3 to 8 428 V
Cathode R = 1K

[jojokeo]

calculating numbers.

that's fun, it's the formulas that suck 

I'm with Tubenit, I have a PP self biased ~320plate, 250 ohm common cathode R setting at 20.5vdc running 82mA or ~~13W per tube

I just finished another 6V6 scratch build and with 378v at node A, etc. - the standard 250r cathode resistor had me red plating at about 130% Pdiss (if I remember correctly?). After several more adjustments I ended up with a 330r (measured cold appx 322 or so?) there and got me down to 13.3w Pdiss (at 95%) and cathode voltage at 24.9v.


*I just mentioned on another thread: hook up your filament CT to the power tube's cathodes and get some free elevated heater action... and congrats on your build success.

[pdf64]

If the idle HT is more than about 400V, cathode bias becomes problematic. Due to the bias voltage rising as signal level increases.
One way around that, but still avoid the bother of a fixed bias supply, is apply fixed bias to the cathode, by means of using a zener diode to fix the cathode voltage at a suitable idle level. So 40V of beefy zener. Power output and ‘fullness of tone’ at high power levels should increase noticeably   

[shooter]

6v6 pin 3 471 V

now that's just showing off! 

[pdf64]

Hopefully your build has a nice low hum level, but even if so, I don’t think that the rationale of separating signal and power supply 0V returns is valid.
Consider that cathode current must include all of that stage’s signal current 
Maybe you’ve misunderstood the concept behind having separate power amp and preamp 0V common points?
Have a read of http://www.valvewizard.co.uk/Grounding.pdf

[nandrewjackson]

Pdf, you can disagree about the "rationale" of whatever you want to I guess. I built it that way to avoid hum. It does indeed have low hum.


Almost all of my builds have been point to point, with ground connections everywhere.  Terminal strip mounting screws, backs of pots, 1/4" jacks,  etc.


About 75% have had low hum. That 25% that didn't was A BEAR to track down the problem. 


This method, rational or not, is a new method I'm trying.
This build is point to point like the others, but
This one has 1 main ground to chassis, and 2 ground buses.  That's how IT IS, rational or not. And you have already "disagreed with merlin" earlier in this thread, so which is it? Are you pointing me to study valve wizard, but also to watch out for the things you disagree with?




As for the bias, and high voltages, I may switch it over to fixed bias. I may do zener diodes on the PT center tap. Like I said earlier,  I tried to get a lower V PT, but they weren't available and this one was an impulse buy.  I may leave it like it is, who knows, I'm not a fortune teller.


Thanks for the positive comments,

[shooter]

fwiw;
if you're not an instant gratification type
I like this PT from Edcor;
xpwr050-120

into a fwb, lots of filter caps, under load, it lands ~~ 340vdc.  works nice for most any small bottle PA tubes, nice small footprint, reasonable $$, and cute 
downside, 6 week delivery

[nandrewjackson]

5u4 rectifier installed. Voltages now at acceptable levels.


Updated schematic picture link


https://ibb.co/r0WTPFj


For those that dont want to click an external link:


Power tubes 2-  6V6
Pin 3 to ground 409V
Pin 4 to ground 412V
VDC across 470R shared cathode 30V
Pin 3 to pin 8, 380V


Preamp triodes
1: pin 1, 172V   pin 3, 1.1V
2: pin 6, 261V , pin 8, 2.7V


Cathode follower
Pin  1 and 7, 176V, pin 3, 1.1V
Pin 6, 291V, pin 8, 177V


Reverb driver AX7
pin1, 316V , pin 3, 2.2V
Pin 6, 184V, pin 8, 1.3V


Long tail Phase inverter
Pin 1, 265V
Pins 3 and 8, 31V
Pin 6, 248V


It sounds like a Marshall based circuit  and the voltages are more in line now.

[shooter]

my supper fingers sticky calculator says 12W a tube

play it like you stole it! 

[jojokeo]

5u4 rectifier installed. Voltages now at acceptable levels.

Based on info available I have Pdiss at 75.6% = 10.58w


*If you want accuracy - before taking voltage readings it is advisable to make accurate cold resistor measurements of your cathode and screen resistors. You cannot/should not simply go by their listed values. Then always allow at least 5-10 minutes to get hot and settle prior to measuring. I would not round low voltage & resistance measurements such as your cathode resistor and voltage, pi tail readings, etc.