Pentode input 12AB5 cheap lightweight amp.

[printer2]

This is an amp that I am currently building. I wanted to make a lightweight amp that could be loud enough to jam with, not that that will happen any time soon. I have a bunch of 12AB5's which are 9 pin 12V equivalents of the 6V6 intended for automobile radios. I have a 1760H output transformer (6.6k 20W) and a 186D120 (120V 0.250mA) by Hammond. The power transformer is not intended for tube circuits and only has the one winding which is center tapped. For the heaters I am going to use a switching supply from a wall adapter. Going to try a pentode up front and see what all the fuss is about with the morph control.



This is the working schematic and will adjust values as needed. The choke is a power transformer which I am using the primary only, will it be adequate, soon to find out. The tone stack will be switched to positions before and after the second stage. I decided to go with a Mosfet follower after the pentode so that there would not be that much loss when the tone stack is loading it down. I also decided to go fixed bias. The amp will probably be going in a 5E3 sized cab. Board partly done with some cheap chassis for future projects made from the lid of an old desktop computer. Using aluminum on this one.



Might go with a 12 or 10, weighed the major components and an anticipating a weight of around 25 or 20 lbs respectively.

[John]

Going to try a pentode up front and see what all the fuss is about with the morph control.

FWIW, I didn't find it useful on mine (I'm running a 6sl7 into a 5879). I did however, find it very useful to make the .1 screen bypass cap variable and adjust the gain that way. Also from Merlin's book in the same chapter I think. But regardless, I'll look forward to your build progress! 

[printer2]

Just in case I do not like it I put enough 'turrets' by the input to change over to a 12AX7 and do a Plexi input.

[printer2]

Just have to wire up the input jacks. I do not have the power supply resistors in as I want to adjust the voltages to taste. Hope it all works.

[printer2]

Well did I really think it would fire up first go round without problems? No, but it would have been nice. Not much hum out of it, power supply voltage dropped quite a bit. One tube has a nice red spot on the plate. Adjust the bias, up to -38V and still can not stop it. Swap out the tube and try to find a tube better balanced to the other one. Still two high on that side, swap tubes between the sockets, same socket running too hot. Bias voltage high, check at the pins, 0V on the offending tube. The bias voltage was shorting out on one side of the dual pot. Bend the terminal and all is fine now.

No signal through the input tube. It would help if I had a cathode resistor don't you think? Left the spot open so I could experiment with different values. Add resistor. Dropping resistors for the filter caps are too high, knew they would be, put in values to get closer to the voltages I want. First section voltages still not right. Might as well add a ground to that filter cap, otherwise it is not doing a heck of a lot.

Still need to lower the output tube bias range. Thought I would be running at higher voltages, might be hard to hit 15W. The preamp section does not have much gain, not much more signal than what is going into the input. Swap out the input tube, no difference.

On the bright side nothing melted. The switching supply is holding the heaters at 12.3V. A signal can go from input to output. A good way to start debugging tomorrow.

[printer2]

Getting funny voltages on the input pentode, when troubleshooting a new build and you do not know what is going on simplify the circuit if you can. I disconnected the Mosfet, still no good. Hit myself on the forehead time, I was meaning to use a 6AK5 (12V version) but was could not remember the tube number. Decided to stick in a 12AU6 instead. Forgot the 6AU6 does not tie the suppressor to the cathode as the 6AK5 does. Jump the pins together and wonder of all wonders, there is my gain. It did not seem to matter switching the tone stack before and after the second stage, might just forget about the Mosfet. The simpler the amp the better.

Since I have the oscillator and scope out I might as well make my way through the amp. Adjusted the PS voltages some. Second stage seems to be working fine. On to the PI, can get 80V clean out of it, I'll have to decide on that later. Drop the input voltage to the power tubes to get a clean sine wave, barely any AC voltage on the plates. What the heck is going on? Poke around some, voltages seem right, got a little increase in output moving the output jack. Play with the jack, have the switch  shorting in case the jack is pulled. The switch seems alright, BOY THAT IS LOUD! Well what do you know, the output jack was shorting out.

As much as I appreciate a good sine wave my guitar is within arm's reach. Turn it up and see what it can do. Sure is loud, going to have to grab another speaker as my little 8" test speaker is not liking its current job. Well the amp works. Now on to wasting away countless hours swapping parts and deciding what sounds best.

[printer2]

Don't think much of the morph control. On the fence regarding the pentode in the first position. I will try it again later, other than a lot of gain I am not getting a lot out of it. Found the tube I thought I was going to use, 6V heater, not worth the trouble.
 
 Thinking maybe going with a 12AX7 instead.

[printer2]

Sticking with the pentode. Played around with some resistors, fed it to a WGS ET-65 that I removed the dope on the cone, has a rip-your-face-off kind of sound. I see some potential.

[printer2]

Really drops more voltage from the power transformer than I expected. Open circuit the Hammond put out around 138V, rectify it and double it  without a load you get 386Vdc, I was thinking it would drop into the 340V range, more like 310V driving the amp hard. So with 310V and a 6.6k transformer that gets you, 310*310 / 6600 = 14.5W. You think I would be happy with that call it a day. But the output transformer is a Hammond replacement for the Deluxe Reverb and is rated for 20W. I want my extra 5W. Will not make it a whole lot louder, just make me feel better.
 
 So I rearranged the power transformer. It is a 120/240V to 120V transformer. If instead of hooking up the two primary windings to the wall I put them in series and hook up the 120V secondary to the line I get 208V AC out. This is because Hammond winds their secondaries a little hot, recall I was getting 138V out, so with the 120V going into it I get less than the 240V you would get if transformers were ideal devices.
 
 So take 208V AC, rectify it which gives us 291V, this is going into a doubler so multiply by 2, 2*291 = 582V dc. That is just plain scary. But since I had some significant losses hooked up normally, what the heck let's try it. I have a variac and I can stop turning up the voltage if there is too much voltage. Keep in mind my capacitors in the doubler are 250V so I should be good for 500V.
 
 Well turned it up and without the tubes conducting it got up to 500V. Which is ok for the output section, will have to look at the preamp capacitors later (I forgot about them, I was just watching the output tube current in case they were about to melt). When they started to conduct the voltage went down to 420V, 415V on the plates. So let's see what that should give me for power. The tubes do not swing down to 0V, maybe down to 40V, so 380V. 380*380 / 6600 =  22W. With transformer losses maybe up to 20W.
 
 I had the bias voltage turned up all the way just in case, going to have to adjust the resistor values as I had 30mA per tube at idle. At 410V*0.030A = 12.3W. The tube plate dissipation is 12W, say 2W of that is going to the screen, about 85% dissipation. A little hot for fixed bias but should be fine until I change the resistors.
 
 I really have to order some 500V capacitors to replace the 350V ones. No room to double them up, maybe I can stack them, worth looking at. Other measurement I took, playing at the edge of breakup the voltage drops to 370V, down to 355V with the amp feeding back and gain at a point where more really does not make much difference. All warmed up the plate voltage is now about 400V. No red plating so maybe the tubes will live. Pretty sure I got the most out of the power tubes. Next is to tweak the preamp. and I have yet to add NFB.

[printer2]

I couldn't wait for the higher voltage caps to come in by mail, doubled up on the 350V ones I had. Looks a mess but I am just amazed that I actually got them and the balancing resistor on the board.

[/img]

[printer2]

Finally sparked her up again. Capacitors are functioning fine, half the voltage across each of the pairs. Time to track down some hum. Lot of hum when I put my hand by the input pentode, take some aluminum foil and wrap it and grounding it, helped a little. Then I realized the suppressor grid was not tied to the cathode again. I removed the jumpers when I was soldering in the capacitors and forgot about this one.
 
 It helped a lot again, still a little hum lingering. Then I thought maybe the switching dc supply for the heaters, I thought the heater dc is not tied to the circuit or the chassis, maybe try it to ground? A little concerned about just shorting it to ground as I do not know how the switching circuit is arranged, took the safe route and shorted it to ground using a 1k resistor. No hum. Now that is a relief. So with the master volume turned up all the way and the preamp volume control turned up I just get hiss coming off the pentode.
 
 So now that I have a quiet amp I can do some fine tuning. Well maybe at least put it in a cabinet.

[printer2]

Part of the reason for this build is that I took part in a challenge to make an amp for $100 minus the chassis, cabinet, and speaker. I received a box full of parts by the originator of the challenge in which was the 12AB5 tubes I used in this amp. Rather than rewriting part of the description of the circuit I'll just cut and paste part of what I wrote there. Measured the values of the resistors and updated the schematic. Thanks John in taking an interest.

Some of what I learned during the challenge made it into the making of this amp. I could have done things a lot easier, buying bits and pieces but I wanted to show there are other ways to do things when you are short of cash. Here is the schematic. The nice thing is the 12AB5 tubes can be had for about $5 each, the 12AU6 also, the two 12AX7's I used as I had them but the dual triode's out of Russia are another way to go shaving a few bucks.

The fixed bias supply is a little more involved than usual as I was using a bridge rectifier and the 0.68 uF cap was needed to isolate the dc, the 22k and 8.2k resistors were needed to get the right range. Since I used two 120V windings in series for the secondary there was no reason to take the full voltage, I grabbed the AC across one winding. I still might adjust the tone stack values although they seem to work for now. The switch moving the tone stack before and after the gain and cathode follower stage was to give a Fender Blackface vibe when before and a Bassman when after. The point was to maybe give a little more headroom sonically by scooping out the midrange before the next gain stage. In practice there is an audible difference but, also some roll off of the highs after the pentode, not surprising. There is not a large a volume loss as I was expecting having the stack after the pentode, part of the reason I got rid of the Mosfet.

The NFB is taken off the 4 ohm tap of the OT, the terminal strip I used only had two positions so I only used the 8 and 16 ohm taps of the transformer. When running one or the other I just move over the wire from the output jack to the right terminal. I have all the capacitor, other than the first one, grounds running to the voltage doubler using a bus ground and then to the chassis. The input section capacitor ground is returned through the input jack which is grounded on the chassis. It is dead quiet. One side of the heater supply is also grounded to the chassis. I had a hum that I could not find the source of and then thought the heaters were just floating.

The Morph control, the 1M pot on the input pentode is subtle in its operation. It gives a little compression to the sound when dialed in. If I did not like the sound of the pentode up front I was going to use this pot for a volume control and stick a 12AX7 up front and convert this amp into a Plexi. I think that about covers everything.



The updated schematic.

[John]

Cool. I haven't commented, but have kept up. Cheers!

[printer2]

Well came in at 24.4 lbs, still need a handle. The standard back panel off shot.



And the sort of front-top shot.



I want to live with it a while before I label the controls in case I decide to do something else with the input. The amp worked fine on the bench but in the cabinet I had something that sounded like a bad connection. Bad output jack for one, did not think these second rate jacks were as bad as they are, no wonder I got them cheap. After that still had some crackling. Poking around on the joints I saw this.



I have a couple of resistors in the power supply filters that are unsoldered and just wrapped around the turret. I was not going to solder them in until I was sure I had the voltages I want. Guess I can't do that. The above joint with the dropping resistor wrapped on the screen resistor lead was continuously arcing. I have seen arcs in equipment before but not one that sat there lit up like a LED. I soldered all the resistors in place for now.

And I had it cranked for 10 minutes tops. I am going to have to roll off some bass. With the bass control at zero I had enough playing clean and distorted with my Tele. I removed the NFB, I did not have very much but the amp sounded more lively without it. The sound was ok but something was missing, no sparkle. Then I changed to a better guitar cord. Much better.