AB763 Deluxe Lite Build

[HommeMarrBuckley]

Well, switched the secondaries with the heaters and put a jumper across the (+) lugs of plate cap.


Turned it on--then flicked standby---limiter bulb lit up (didn't previously)--then a bit of smoke--shut it off quickly.
so perhaps I had the secondaries and heaters correct in the first place.
I just don't know with standby and plate cap wiring.
It's wired how I wrote--just now with that jumper which shouldn't matter.
Considering scrapping this--learning the hard way---without experience should go with a proven build.

[HommeMarrBuckley]

Alright, haven't given up quite yet.  I believe my plate filter cap wiring is off.
Just asking anyone for help here.
If you look at this schematic of the power supply and to far right (2 70uf and 2 220k resistors)--that's what I want to try and replicate with my cap can.
http://www.lynx.bc.ca/~jc/63PwrSupply.gif
Im using F&T 32/32 cap can as the plate cap and have one 220k resistor (but have extra if I need second)
Could someone write or scribble pic how I should wire my can to replicate that?
I know one (+) lug meets rectifier board and goes off to standby

[sluckey]

If you look at this schematic of the power supply and to far right (2 70uf and 2 220k resistors)--that's what I want to try and replicate with my cap can.

Can't be done using only that cap can.

We covered this earlier. Back up and read this whole thread again.

[HommeMarrBuckley]

Can't be done using only that cap can.
We covered this earlier. Back up and read this whole thread again.

You're referring to your last comment in post#38, correct?
I honestly was thinking about that but didn't realize way I intended was not possible.
Alright, I am going to try and put your comment for using the can as both plate and screen.
Thank you for pushing me in right direction.

[HommeMarrBuckley]

That's not what I had in mind. I was thinking use one 32 cap for node A and the other for node B. OT would connect to cap A, choke between cap A and cap B, and screens to node B. This would be a clean, straightforward way to utilize that dual can.

I follow

Can't be done using only that cap can.
We covered this earlier. Back up and read this whole thread again.

but now am not sure if you meant that I can't use that can solely as the plate cap or that I can't use it all
If I can use it:

1---where does my rec board (from diodes) line connect? go straight to standby?


2---then would wire from node A go to other standby lug?




3---Or can't use the can and go with 5 20uf caps? 2=plate 1 screen 1 PI 1 preamp

[HommeMarrBuckley]

Well,
After spending the last few hours unhooking the cap can and then creating a 2 20uf plate cap board and hooking it up---only to get dejected to see the lamp limiter bulb keep a dim light just like before--I didn't know what to do other than to look around.
I used Moogami shielded cable and it's very frustrating to use.  The hot lead is so fragile and thin and tends to break so easy if you turn the cable.  Things are hella close but couldn't see anything touching--lead to lead or to chassis.  Then I looked at the power tubes thinking that I needed to have the 1 ohm resistors jumped from pin 8 to pin 1.  That was a no but then I saw something.  I forgot something.  I didn't put in the 1.5k resistors across pins 1 and 5.
I can't believe it---If this solves the lamp limiter short--I will be more than ecstatic--but something tells me there will be something else--these pt wires--ugg but the lamp light is not that bright.  If it does work----l might have to question reinstalling the cap can. Because it's better quality than the electrolytics I put in and they are in there super tight and close to many other components.

[Willabe]

........but the lamp light is not that bright. 

There might not be anything wired wrong/shorting.

Depending on the the bulbs wattage and depending on how much current the amp draws the build will stay lit to some degree.

The lower the bulbs wattage and the more current the amp draws the brighter the bulb will glow even though the amp is wired correctly.

Try this, take a known running amp that is as close to the same power/wattage as the amp your building right now and use the light bulb limiter on it.

That way you'll get a feel for how bright the bulb stays after the 1st in rush current surge that charges the B+ filter caps and the heater current surge. The heaters draw something like 5 times(?) the current when they are still cold. As they heat up they draw less current until they settle in at their spec sheet current draw.

[HommeMarrBuckley]

In this case, there's something.
I let it run for a bit, but the limiter is lit more than it should be.
I was measuring the pre amp cap and got 439 volts and then I started smelling and then seeing smoke.
I shut it off quickly and saw that I fried the 470 ohm 1 watt resistor. (chopstick on right is pointing to it--grayish)
It's part of bias supply---on layout it goes to bias tap--I have it (purple wire) going to top half start of rectifier board (chopstick on left is pointing to it)


Any thoughts?


Also, in terms of wiring
red ot <---and --->choke   ---------->standby lug


|> -----|> ----
                       \
   rectifier ------X--> leg of 220k resistor ---> +20 uf cap ---> +20uf cap ----->standby lug
                       /                         
|> ----- |> ----


choke <---- B+ cap ----> pins 6 and 4  of V3 & V4




Down and tired but not out

[Willabe]

If it started to smoke then it's drawing too much current.

You could try this; take the power tubes out and disconnect that purple wire.

Now with the power tubes out power up the amp with the lamp limiter and see if the bulb goes dim.

If it goes dim then reconnect the purple -bias wire and work on the -bias supply. Something is pulling too much current through that 470R, you have to find it.
 

[HommeMarrBuckley]

I don't have the power or pre tubes in.
Was just going to check DC voltage with limiter at caps and tube pins.
 

Wonder if I should add 2 more diodes to rec board?

[sluckey]

You can't use a 470Ω for that resistor because your PT does not have a bias tap. Replace it with a 100K 3W. Both of those bias caps are probably toast now from the several hundred volts negative voltage that they have been exposed to. They have certainly been stressed.

[HommeMarrBuckley]

I have a 2 watt 100k.
Should I hold out and get 3 watt or higher?


For the bias caps--I have some sprague 50uf/50v
Would that be satisfactory or need 100V or more?

[sluckey]

What is the voltage rating of the bias caps in the amp right now?

[HommeMarrBuckley]

2 47uf/160 V


Don't see any bubbles but what do I know.

[sluckey]

Try'em. They may still be good.

[HommeMarrBuckley]

AHHHH!!!! #$%$*#'  RIGHT!!!!!!!!!!
No more lamp limiter issue!!! And numbers seem good!


After using the Limiter--these numbers are from plugging right into the wall.


Caps                                    V1                    V2                      V3                       V4
Reservoir  454              Pin 1   445           447            Pin 3   453                  453
Screen   447                  Pin 6   446          446            Pin 4   452                   452
PI  446                                                                        Pin 5   -26.3                -26.3
Pre  443

I believe my limiter staying lit issue was my prior filter cap layout.  I did not have them properly in parallel and I was using a 220 k resistor that I believe that was not in order, as well.  I should know better because I know that orientation and I don't know why I added extra nodes and strayed.


I went with (2)22uf for the reservoir caps which I believe equals about 44uf/900V capacitance and didn't use a resistor.  The Robinette site is very good for this amp.  I was reading it this morning.  I lied to myself earlier (in the post) and have the MV put in so I am curious to hear how it is when I add tubes.  I will go forward with biasing and to not exceed 12 watts for each 6v6.  Time to clean up and and enjoy this is a bit.  Will be fully satisfied when it's fully working. 

[HommeMarrBuckley]

Ok turned on with tubes and then switched standby to operate and got Hellish loud shrilly/squeal.  Quickly shut off.
Hopefully this is just OT issue.
Should I switch the OT on power tubes? or can I wire negative feedback resistor wire differently?

Steve, I remember you did something to your jack instead of reversing the OT wires--dunno if it can apply here, would be great because it's easier fix.
Bright green is NFB with green ot, then jumper--->Black OT


Will put NFB with black OT on tip, green OT on ring with jumper going to ground.
Believe this is correct.

[HommeMarrBuckley]

Ok, switched the NFB wire on the jack and turned on and no squealing so believe that is solved.


But at a standstill now, went forward with biasing.
I got 457 on pin 3 of V3.
Went to check mV on pin 8 and got 0 which seemed ok because bias pot was turned the whole way counterclock wise (to left).
Slowly started turning the bias pot and still nothing, 0mV, if full sweep is 100%, I turned it to about 30% and still got 0mV.
Heard a bit of tiny crackling.  At start, power tubes were dimly lit, as I turned they got bit brighter, plugged guitar in but no sound.
I know I damaged that bias resistor prior---so nervous about continuing to turn the bias resistor if something else in that path got damaged.
Is there a way I can check bias supply to see if something else got fried?

[HommeMarrBuckley]

Ok--Update
I checked the bias pot and it continues to sweep to 9+K I checked the resistors around the bias on the board and they were correct.  to further check the bias pot, I checked pin 5 on V3 and was -26, (bias pot completely to the left) I turned it a bit to the right and checked again and pin 5 was 25.2, so I know I have it set correctly for maximum negative DC.
So, I turned switchby on and checked the bias and it read 49mv   so I sure didn't turn it right to increase!
I had guitar plugged in, so strummed it and sure enough--sound--but super, super loud, barely had guitar volume up and normal volume on amp was probably at like 2--but there was sound so good sign!  Just didn't expect it because it was dead quiet. 
Oddly enough, I didn't mess around too much for fear of damaging the output tubes--but they didn't appear to be super red or overheating--but shut it off.
So Im thinking I need to add a resistor somewhere in the bias supply or those 2 bias caps aren't fully functioning.
I know I am very close.
Going to take a break on this for a bit (but not for long )---been at it almost non-stop for past two days and been living like caveman.  Still will be in my head though because honestly believe I am so close.
 Anyone's thoughts, suggestions, troubleshooting ideas?

[sluckey]

You need to be able to adjust the bias voltage on V3 pin 5 from about -25 up to about -40. If the max you can get is only -26 then you'll probably burn up those 6V6s.

That 100K resistor you put in the bias circuit is the range resistor. Lower the value to increase the voltage on V3 pin 5.

[HommeMarrBuckley]

I think I will try a 470Ω


 Im thinking a 2W/10K to start.

[sluckey]

NO!!!!!!!!!!! Try an 82K.

[HommeMarrBuckley]

Ok, I have 82K but not in higher wattage. 
Should I wait until I get a 3-5 watt?

[sluckey]

82K may not be what you end up with. May even need 68K, or 62K, or 56K, or 47K. So I'd even try a 1/2 watt. When you find one that gives you the range of voltage that I suggested, then consider getting a 1 watt, or 2 watt, or 3 watt to use permanently.

You do not need any tubes plugged in to determine the correct resistor. Heck, your 100K may already be what you need. You say you get -26v with the pot turned all the way to the left. What do you get with the pot turned all the way to the right?

Once you can adjust the volyage on V3 pin 5 between approx. -25 and -40, then put the tubes back in and set the bias pot for about 25mV at your bias test points. Then measure plate voltage and calculate idle power.

[HommeMarrBuckley]

You say you get -26v with the pot turned all the way to the left. What do you get with the pot turned all the way to the right?

Yes, I thought for some reason I had the Hammond pot wrong--but no, when I started to turn it right the negative voltage went down, -25 to -24 and so forth.  I didn't think it was a good idea to continue turning it because I figured it would just continue to decrease. 
I might check it again but everything that happened seemed that's the case.  High mV bias reading, super loud.  Only thing not sure---was the tubes, but then again I never saw tubes burn up either so perhaps they were getting hot. Just didn't wait around to see that possibility.  I am using JJ 6v6S they each say 17/2.3 on box--guess that's something with matching--never read up on that.


Good to know I can do without the tubes in and I will experiment with different resistors.

[sluckey]

You say you get -26v with the pot turned all the way to the left. What do you get with the pot turned all the way to the right?

Please answer.

[HommeMarrBuckley]

It slowly went down as I turned it right----I don't understand, as I continue, you're saying it's going to reverse at some point and the negative DCV will increase?
But if you believe and say it won't do any damage, I will go ahead and turn it the whole way to the right.
The tubes are out.

[Willabe]

He want's to know the -bias dcv at 1 end of the -bias pots rotation AND the -bias dcv at the OTHER end of the bias pots rotation. In other words, he want's to know the full -bias dcv swing available from the bias pot set full up CW to full down CCW.

Please post BOTH -bias dcv's.

Remember to do this with the power tubes uninstalled/removed.

[sluckey]

You don't answer my questions. I don't answer your questions.

[HommeMarrBuckley]

Ok, so I believe with this full sweep measurement, this will help calculate the new range resistor.
full left -26.55
full right -14

[sluckey]

Do you have a proper range of bias voltage yet?

[HommeMarrBuckley]

I haven't tried to swap the bias resistor, yet.  Will get to it, just had some late hours and came home with no energy.  Hopefully, will have some time tonight, I dunno.  I will start with 82k and report when finalized.
When youre done soldering, how much time you let cool/sit, before you power up.  I usually give half hour or more if I can.  Just curious.

[sluckey]

Immediately! Sometimes I don't even turn the power off.

[HommeMarrBuckley]

Alright, Here's what I got with the 1/2 watt resistors I had on hand.
                     V3 Pin 5
82k                 -31.3
68k                 -35.7
56k                 -43.4


Go with 56k? or Perhaps I should use a 62k?  Im thinking 62k might get it closer to -40

[sluckey]

Looking for a range of voltage when you turn the pot.

[HommeMarrBuckley]

Alright,
So, I should have been looking at the range.
Am I taking that reading on pin 5 or the 1 ohms?  Or doesn't matter?

[HommeMarrBuckley]

On pin 5, and looking for a range of -25 to -40


56k   -23.09  -  43.4
68k   -18.9  -  -36


Didn't bother with the 82k because of initial -31.3 reading


I just wish I had a 62k to check.
Thinking maybe to order 2-3 watt 62k.

[sluckey]

56K looks pretty good to me. That should let you set the test points to a safe 25mA and go up or down as needed.

[HommeMarrBuckley]

Alright, sounds like a plan, awesome!
Will update when I get ahold of them.  Can't wait to install and put tubes back in and give this thing a whirl.  Good vibes coming.


I was also wrong--this has been done
http://el34world.com/Forum/index.php?topic=20460.0   (great audio and musicianship)
http://el34world.com/Forum/index.php?topic=9891.0
Perhaps more, but what I have discovered in working on this.


Only thing I am not sure is--I had my reservoir cap wiring all messed up--which wasn't getting me anywhere---but wanted to know If I could I have used the cap can wired in parallel as the reservoir cap?
Thanks

[sluckey]

I was also wrong--this has been done

What are you talking about?

but wanted to know If I could I have used the cap can wired in parallel as the reservoir cap?

Absolutely. But you can't wire that can in series like you asked about.

[HommeMarrBuckley]

What are you talking about?

I was whining about trying to do something above my comprehension---it was all there but I couldn't figure it out and couldn't find something to reference.  It wasn't until I saw that euro version and started to look at Fender layouts that I realized what I was doing wrong.

But you can't wire that can in series like you asked about.

Yes, that makes sense now.  I was just thinking of putting it back in place but couldn't remember why I asked about in series but I just remembered it was related to that link that showed 70uf in series.  The robinette site mentioned not to exceed 60uf capacitance for the reservoir so regardless, the can in parallel would not be wise and in series would not work and be too low anyway.  Just have to use it in another build.

[sluckey]

The robinette site mentioned not to exceed 60uf capacitance for the reservoir

That only applies if you have a rectifier tube. You can use any value cap you want with silicon diodes like you have. Marshall is fond of using dual 50µF cap cans tied in parallel for their filter caps. Your dual 32µF can will be fine. Look at this...

     http://el34world.com/charts/Schematics/files/marshall/Marshall_JCM800_2204_Lead_Series.pdf

[HommeMarrBuckley]

Well, my resistors arrived today so I installed the 56K and put the tubes in.
Here are the results:
V3 Pin3
442V      x     21mV = static dissipation wattage of 9.282


V3 Pin3
429V      x     21mV = idling static wattage of 9.009


And it is working!....
but this thing is really loud---I mean I could not put the volume past 2 or 3
I couldn't get to any grittiness or any kind of breakup because it was so LOUD.
Also, the Master Volume doesn't work--if this mod applies I would have to attribute it to my wiring.
I think the Master Volume pot wiring is ok but Im thinking about the pot wires and where they meet the 220K resistors on the board.
Im having them connecting (on layout) at top of 220Ks where bias wiper meets as well.
I believe I need to have them connect at other end of 220Ks where they hook up with .1 coupling? caps.
When I compare what I have to that mod, that's one thing I see that I believe to have done wrong.

[HommeMarrBuckley]

Awwwwww Yeah!!!!!


Well, that was it.  Connected to the other end of the 220Ks and rechecked bias just to make sure.
432 x 22.5 = 9.72
425 x 22.5 = 9.56
Push Pull pot down--No MV
Push Pull pot out--MV works!
Pretty cool--so, of course, I cranked the regular volume to about 8 and just slightly put MV bit past 1 and it was totally rad.
Great amp for bedroom or to play out.  Can't say enough about the master volumes in these past two amps--plexi and this.
Now, on to make a cab to house this.
Many thanks Steve and forum members!

[Willabe]

Hehehe, alright! Got her up and running.   

[sluckey]

Congratulations! Everybody needs an AB763. And now you are the 'go to guy' for AO-43 conversions.