Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: dwp on May 21, 2010, 10:26:02 pm
-
Hello All, For additional boost without blowing it's top, can I take my EF 86 channel, (rotary tone cap control) currently connecting at the PI, shared with a plexi preamp, (via two 220K's to pin 2) and move it up to the plexi preamp for additional gain? (say via an additional 270K to pin 2 of V2) Or additional suggestions, please.
I was hoping the single EF86 channel would be all romp, it is not....played dimed is about the tone, any incriment less is alittle disappointing, thin. not worth the space it is taking up at present IMO. I like to ask before I get adventurous. My record of no smoke/flares/fires/DAMAGE is holding to date.
Thanks in advance, dwp.
-
Got a schematic (that you could post or link to) of the amp you're working with?
-
Thanks geezer, I do not have a schematic. (not versed well enough with the .sch) but.......
Is a standard plexi/5f6A build. then I took the EF86 circuit hosed from ceriatone,
http://www.ceriatone.com/images/layoutPic/marshallLayout/36WEF86NewCeriatone.jpg
the EF86 end only. then volume wiper from EF86 channel to pin two of the PI/V3. .022 cap, two 270K resistors, one for plexi treble wiper, one for EF86 volume wiper. This works good with regards to channel independence, and the plexi end sounds great as it should. No noise issues anywhere, just not enough oomph from the EF86.
I would like to have the EF86 channel HOT!!!!
thanks again, dwp.
-
I would try to sneak the EF86 in the middle of the other channel. Maybe after the first gain stage. It's a Pentode. They sound good when pushed... Seriously, try it.
-
Thanks Loosechange, that's where I was going with this.
I want to confirm: move EF86 Volume wiper from PI, (V3-2) to V2-2, and add a third 270K resistor along side the two from 5F6A Volume wipers?
My board is same layout as Hoffman 5F6A/Plexi. (septin the EF86 end) Will be going through the plexi tonestack, but that's good, yes?
thanks again, dwp
-
I was thinking cascading the 5F6A first gain stage into the EF86 and then into the PI.
You're thinking parallel. What you say will work too.
-
I was thinking cascading the 5F6A first gain stage into the EF86 and then into the PI.
Similar to what I did with my T-Lite 5879 OD.
With respect, Tubenit
-
Tubenit's got the plan right there! :grin:
The EF86 is just a bit higher gain than the 5879, but same basic idea/tone.
-
thanks tubenit, your drawing/building is lightyears ahead of me.
So I took the Ef86 channel, kept the rotary tone circuit, and just inserted it at V2-2 instead of PI, via 270K resistor, and no ill effects, works great. I have a rotary pre-set so to speak now, then one tone stack serves all. The tone is actually incredible now, I've only had last evening to play with it, will see what develops after a harder/louder tryout.
I'm going to study your switching somemore, that will be a great feature to have a relay to switch the EF86 in and out, stomp lead channel. For this build, I already have the segregated control layout, chassis by Ceriatone, very nice BTW, welded seams, etc.
This is the 4 EL84 build (frown) that I finally did punch the 9 pins out on, added 4 Octal for 4 6V6 or two 6L6/KT66. Haven't bought a Quad of 6V6 yet, but the 6L6's sound amazing. lifeless has come to life......
Thanks again all for your input, dwp
-
The dpdt is a mini-toggle. Doug Hoffman uses something similar on his HotSwitch mod. The Dumblish guys use something similar also but with DPDT relays.
Nice thing about the T-Lite is that I can set the clean and OD channel for exactly the same volume OR have the either be slightly louder. It works great. I love the amp & the 5879 gives a nice smooth OD.
With respect, Tubenit
-
Hey still out there Tubenit? So I'm really pleased with my set-up, feeding the EF86 through the 5F6A tonestack. The Tone Cap Rotary Selector allows a tele bridge to be colored alittle darker, a strat neck to be colored alittle lighter. Your set-up feeds a first stage through the EF86, can you offer alittle insight on the different result? Do the additional millivolts on the signal tend to make the EF86 less stable? Speaking to it's inherent noise problems only. My GEC is titties. (can I say titties?) no squeeling, even loud, loudest, was surprised. My 4 X 10 Super Reverb Repro (have been to the mountain and back) is fixing to go tweed w/EF86. Am going to switch a single set of jacks, am just curious about whether to copy mine, EF86 to V2-2 or try yours?
Do you use isolated tube sockets? I'm trying to find another one like this, should have bought more when I had the chance, Pacific TV in BC. they're gone now, forever. (the sockets I mean)
Thanks again, dwp
-
Never used an EF-86. If you've found a tone you like, perhaps leave it alone?
I use a silicon hot pad (like for cooking) & cut it up to use as a socket damper. Works great.
With respect, Tubenit
-
Quest for tone, yes. I do like, yes, three-four days in a row so far. The SR is much hotter iron, is my concern. I will experiment sending first stage through the tube (EF86) retaining the rotary selector, and/or direct to V2. Am just realizing your 5879 isn't an exact equal, different pinout. Will start a new thread with the new build.
Thanks again, dwp
(found my receipt, I did buy 2 of those sockets back in 6/2008, if I sent one to someone, please ring in, I've been through all my spots, no idea other than I know I have only one EF86 build, my first. Old age setting in, tattered mind)
-
This may be too much theory, not enough practical outcome, but...
You calculate the gain of a pentode differently than for a triode. The first tedious requirement needed is the transconductance of the tube at the operating point used. Once you have that, the gain of the pentode stage is the Gm (in mhos) times the a.c. plate load resistance.
Say you have a pentode with a Gm of 2000 micromhos and a 220k plate load resistor. Also say the following stage has a grid reference resistor of 1M. The a.c. plate load is not just the 220k, but because of the coupling cap, also the 1M in parallel with that. So that's 220k ll 1M or about 180k for our example. 180k * 2000 micromhos = 180k * .002 = 360. Much more than a single 12AX7 stage, but not as much as cascaded stages.
A helpful thing to remember is the meaning of the mho (or the british way to refer to it), which is current/voltage; that's the opposite of the way you'd write Ohm's Law for resistance. So 2000 micromhos is also said 2mA/V. 2 milli * 180 kilo = 360.
The a.c. load for all triode and pentode stages is always smaller than what you see for a d.c. loadline, due to the following stage's input impedance being in parallel with the load resistance. Except in a couple of notable cases... namely, when the following stage has a bootstrapped grid resistance like the split-load inverter, long-tail inverter and some variations of the cathode follower. We don't need to calculate the input resistance of these stages fully, just know that it is likely much more than 10 times the size of the plate load resistance. That means that when placed in parallel, there is almost no reduction of load resistance at a.c.
The implication of this is that as long as a pentode couples directly into the bootstrapped stage, and we allow enough current to flow through the pentode, you can use a large vale of plate load (like 500k, possibly more) to get even more amplification from the pentode.
There are limits, and a big trick is allowing enough current through the pentode, and ensuring there is sufficient output voltage capability to use the gain available. But it is worth knowing. For example, you could use this trick to get full output from a pentode into a phase inverter to output tubes with an input in millivolts to the pentode. The drawback is that you can't put a tone stack between the pentode and the phase inverter for the full potential gain. But you could have a triode stage and a tone stack/volume control, another triode if desired, and still be able to drive the snot out of the output tubes.
-
HPB, thanks for the great explanation. You lost me shortly after "calculate"....(lol)
I'm landlocked still in copy and paste so to speak. But the theory is rubbing off, one year at a time. I'm still mastering switching and tonestacks. And the ability to tame......
I look forward to your review on my next build, will be tackling the Super Reverb conversion very soon.
thanks again, dwp.
-
I use a silicon hot pad (like for cooking) & cut it up to use as a socket damper. Works great.
With respect, Tubenit
I can see how you cut out the flat disc part to mount the socket onto but how did you make the part that wraps around the tube and/or shield?
-
I don't remember, but Tubenit might have just cut a strip and glued it to the shield with RTV sealant. Like high-temp gasket sealer from an auto shop.
HBP got it right. Tubenit
-
> You calculate the gain of a pentode differently than for a triode
No, same way. You look in the amplifier tables the tube-makers supplied!
This may take some digging. The amp tables are often not in the data-sheets, but a separate compilation in the back of the manual. Which not all data-sheet compilers bother to scan.
And if you are doing something very whacky, it may not be in the tables (and you should wonder if you are maybe off-track).
5879 in gitar amp.... you probably have nearly 300V available and you rarely have a load a lot higher than 220K (counting everything) (bootstrapped driver perhaps an exception). So you proably want one of the conditions I outlined in red below. Also see http://www.tubebooks.org/tubedata/HB-3/Receiving_Tubes_Part_1/Resistance-Coupled_Amplifiers.PDF
-
Thanks for your reply PRR. Am still confused as ever trying to follow the chart outline. Also the EF 86 is nowhere to be found, slightly different pinout, yes? I've played with the plate voltage, but I don't hear a lot of difference. the output was sufficient directly in to the PI for moderate play, but in to the second stage of 5F6A, it stands up, taller. Much taller.
Still don't get the reason for 470K Marshall at V2-2 versus 270K Fender.
Thanks again, dwp
-
Also the EF 86 is nowhere to be found, slightly different pinout, yes?
http://tdsl.duncanamps.com/show.php?des=EF86
Scroll to the bottom, there's loads of info for this tube.
The larger the load resistor the less voltage to the plate and the less gain of the tube. The lower the load resistor, the higher the voltage to the plate, and the more gain of the tube. The screen resistor is commonly 5 times the load resistor but can be more or less. Lower screen resistor gives less less headroom and larger than this gives more headroom. Start out with 220k for load resistor and 1M for screen resistor. Adjust as necessary depending on tastes and voltages in your circuit. Hope this helps?