|
Post by Peegoo 🏁 on Feb 22, 2023 19:14:10 GMT -5
This is an appeal to you all for your thoughts re: this circuit. It's a kludge of a Matchless Lightning, a Marshall 1974x, and an amp called the T-Rex 18. I don't have voltages on this yet, pending a looky-loo by you folks. It's called a Zippo because it's...match-less, which is the cleverest thing on the schematic. Which ain't much. V1/2/3 are 12AX7 V4/5 are EL84 V6 is EZ81 There's one goof in the drawing, and that is the connectors: there's A to A, B to B, and F to F. No C's or D's. Just three pairs. Don't worry about insulting my intelligence on this stuff because I have barely enough to know which end of the soldering stick to grab. I'm learning as I go. I have a hum balance on the heater supply. Some builders say this is a bad idea, and some say it's fine. If you see anything else on here that I can improve, please let me know. Cheers and thank you kindly.
|
|
pdf64
Wholenote
Posts: 557
|
Post by pdf64 on Feb 23, 2023 3:58:45 GMT -5
Yikes, that’s going to be a beast! I like the elevated heater circuit, I’ve used that on my last couple of builds and it’s a game changer, valves that were previously unusable due to heater buzz are often totally quiet. I suspect that due to the how heater elevation works (saturates any leakage current paths), the humdinger pot wont be beneficial, but then it shouldn’t cause a problem either.
At first look through, I see 3 problems.
The input socket has the 1M grid leak on a switched contact, so won’t be connected when anything’s plugged in. And as the input isn’t shorted with nothing plugged in, it’ll pick up hum.
The screen grids need a better filtered feed from the HT supply. EL valve types have high gm, any HT ripple that finds its way to the screen grid will get amplified and modulate anode current. So I suggest to add another node after the reservoir cap that feeds the OT CT, 1k - 2k2 or preferably a 10H - 20H choke. As per the Lightning. It would be much better for the EZ81 to have its own heater winding. Otherwise its heater - cathode insulation has almost full HT DC voltage across it, which was never good practice even with vintage manufactured valves, but may make current production EZ81 unusable.
Oh, as there’s only a single input socket, the input grid stopper R8 can be 33k rather than 68k. 68k will noticeably increase the hiss level.
|
|
|
Post by Peegoo 🏁 on Feb 23, 2023 4:34:09 GMT -5
Okay, this is a huge help. Good catch on the switched grid leak resistor; dopey move on my part!
I have a few chokes I can use for the OT CT, I was wondering if it were really necessary because some amps don't use one. I'll stick one in there to head off the hum.
The PT I have here does not have a secondary for another heater circuit to run the rectifier...I'll have to get creative to add one for the EZ81. I was also considering diode rectification for this amp. I've used it twice before and it seems to work great. This was a thought because I looked at prices for EZ81 tubes and they are stupid expensive.
Thank you for the look; I really appreciate it!
|
|
|
Post by Peegoo 🏁 on Feb 23, 2023 6:39:52 GMT -5
Any thoughts on the following? This is why I originally went with heating the rectifier with the same supply as the power and preamp tubes: ----------------- Valve Rectifiers Some small valve rectifiers EZ81 / 6CA4 are indirectly heated and can be run from the same heater supply as the other valves in the amp. Most, however, are either directly heated (i.e. the heater and cathode are one and the same) or the heater is connected internally to the cathode. ----------------- From: www.valvewizard.co.uk/heater.html about 3/4 down the page.
|
|
|
Post by Peegoo 🏁 on Feb 23, 2023 8:22:27 GMT -5
I made some changes to the schematic. I'll add a small 12v transformer and step it down to 6v for the rectifier heater. Re: the choke...I think what I'm going to do it use clip leads on the choke to see where it performs the best, and then make the connection permanent.
|
|
pdf64
Wholenote
Posts: 557
|
Post by pdf64 on Feb 23, 2023 9:42:24 GMT -5
Regarding the EZ81 heater supply, sorry, I wasn’t suggesting there’d be an immediate failure, rather that it opens the door to another failure mode, making the amp a bit less resilient. I recall reading somewhere that pretty much all indirectly heated valves used the same heater to cathode insulation technology, which (in standard form at least) was only really good for about 100V. Good mitigation would be to add fusing to both legs of the feed from the PT HT winding to the EZ81 anodes. Which in my view is always a good idea anyway. See the valvewizeard page on fusing. Regarding the choke, if you’re not too bothered about maximising the clean power output, eg for big clean chords, then a resistor is fine. I suppose if you go to silicon rectification, the extra voltage dropped across a resistor to the screen grid node may be beneficial
|
|
|
Post by Pinetree on Feb 23, 2023 10:05:19 GMT -5
Dude.. check yer mailbox.
|
|
pdf64
Wholenote
Posts: 557
|
Post by pdf64 on Feb 23, 2023 12:28:56 GMT -5
That line out may be too hot. The Vrms at the 16ohm output could be nearly 20V, so maybe reduce R34 to 1k - 1k5 for about 1V. Or whatever gets you’re preferred line level.
|
|
|
Post by Peegoo 🏁 on Feb 23, 2023 14:46:58 GMT -5
Dude.. check yer mailbox. I will this evening...I'm out at the shore house. Must remember the asbestos gloves I recall reading somewhere that pretty much all indirectly heated valves used the same heater to cathode insulation technology, which (in standard form at least) was only really good for about 100V. Ohhhh NOW I understand. Makes perfect sense. And I will add a fuse to the HT supply to the recto, even if I stick a bridge there; I know some amps employ it, but it's really cheap insurance. That line out may be too hot. The Vrms at the 16ohm output could be nearly 20V, so maybe reduce R34 to 1k - 1k5 for about 1V. I didn't even consider that...that is speaker level; I'll attach it to the 8-Ohm tap. Seems sort of important if I don't want to blow up any downstream input stages. Thank you sir!
|
|
|
Post by Peegoo 🏁 on Feb 23, 2023 18:29:32 GMT -5
Okay, here's the schematic with changes. 1. I moved the line-out signal source to the 8-Ohm speaker tap; this is identical (same resistor values) to the M Lightning. 2. I added a choke between B+1 and B+2, similar to how it's implemented in the Lightning. Will this serve the purpose of reducing ripple on the screen grids where it is here? 3. I swapped in a diode rectifier. I'm itching to dump all the ingredients into a bowl, mix it up, and shove it in the oven. But I want to make sure I'm maximizing my chances of success. Thank you for your wise counsel, Pete. Anyone else please feel free to chime in...I want to learn as much as I can about this stuff.
|
|
|
Post by markfromhawaii on Feb 23, 2023 20:58:58 GMT -5
I’m wondering if the choke should go right after the output of the bridge rectifier. BTW, what are you using for CAD? Nice schematics.
|
|
pdf64
Wholenote
Posts: 557
|
Post by pdf64 on Feb 24, 2023 4:45:41 GMT -5
The choke Is feeding the phase splitter node, with the screen grids are still being fed from the same node as the OT CT. A new node for the screen grids needs to be inserted between B+1 and B+2. A full wave bridge rectifier is wrong for this application. The way it’s shown, its negative output (ie -350V !) is being shorted out. What’s needed is a 2 phase rectifier, just delete the 2 diodes on the right side. If using typical 1N4007 diodes, as their 1kV rating is a bit marginal in this application, it’s best to double them up in series pairs. Have a looky www.valvewizard.co.uk/bridge.html
|
|
|
Post by Peegoo 🏁 on Feb 24, 2023 9:23:36 GMT -5
I’m wondering if the choke should go right after the output of the bridge rectifier. BTW, what are you using for CAD? Nice schematics. I'm old-school. I've been using MS Paint for many years for this stuff; I have sheets of components already drawn, so I drop them in and connect them with straight lines. It takes a little longer than CAD, but I'm used to it and can work pretty fast with it. I probably should step up and live in the 1980s
|
|
|
Post by Peegoo 🏁 on Feb 24, 2023 9:24:50 GMT -5
The choke Is feeding the phase splitter node, with the screen grids are still being fed from the same node as the OT CT. A new node for the screen grids needs to be inserted between B+1 and B+2. A full wave bridge rectifier is wrong for this application. The way it’s shown, its negative output (ie -350V !) is being shorted out. What’s needed is a 2 phase rectifier, just delete the 2 diodes on the right side. If using typical 1N4007 diodes, as their 1kV rating is a bit marginal in this application, it’s best to double them up in series pairs. Have a looky www.valvewizard.co.uk/bridge.htmlOkay, I'm getting closer, thanks to you. "Peegoo, ya big dummy!" I got it...I got it. If you hear in the news about a widespread power blackout on the US east coast, it was probably me!
|
|
|
Post by Pinetree on Feb 24, 2023 9:27:24 GMT -5
One of the funniest things that ever happened at work.
Electrician over the radio: "Okay try it."
Then the entire shop went dark.
|
|
|
Post by Leftee on Feb 24, 2023 10:10:17 GMT -5
“OK… untry it!”
|
|
|
Post by Peegoo 🏁 on Feb 24, 2023 16:08:23 GMT -5
It's called a "smoke test" for a reason. In this instance, fireworks are NOT a reason to celebrate Lucky for me, I do have a dim-bulb tester, as recommended by many amp gurus. Problem is, every time I go to grab it off the shelf it flashes wildly and makes a WOOT WOOT sound. I don't know why this is.
|
|
|
Post by Peegoo 🏁 on Feb 25, 2023 21:38:42 GMT -5
Okay, coming down to the wire [har!] on this drawing. I think I have everything correct, after following the Yoda-level advice here and doing some more reading up on this stuff. Changes on this drawing below: 1. I swapped in a silicon rectifier, three diodes per leg. Several references suggested two per side like you did, Pete, but added that three are even better because three in series further reduces the voltage across each diode, increasing their reliability. They're cheap, so why not? - Should I fuse these as on a tube rectifier? - With the swap-in of the silicon rectifier, is a standby switch a good measure to protect the cathodes of the tubes? Some references say yes. 2. I added C1a and R1b for further smoothing. Is this not necessary or is it a good measure? 3. I split the power supply to the OT CT and the power tubes' grids with a 10H choke. Thank you Sensei!
|
|
pdf64
Wholenote
Posts: 557
|
Post by pdf64 on Feb 26, 2023 6:35:05 GMT -5
1. I’m an advocate for HT fusing. I suggest a quick blow ceramic 200-250mA, inserted between the blobs at C1b and the choke. Ceramic may be best because if it blows, there’ll be a back emf spike from the choke. I dislike standby, but the DCCF circuit of V2 can cause problems for some (Russian, JJ) valves. Merlin shows a simple and transparent protection mod, just a diode and resistor. I feel conflicted for advising about the EZ81 heater issue now. www.valvewizard.co.uk/dccf.htmlIf you still want standby, he shows a trickle charge mod. www.valvewizard.co.uk/standby.html2. R1a is just discharging the HT caps, in case you forget to when messing around inside the chassis. It’s not helping with smoothing at all. C1a may be overkill for a lower power amp, C1b 30uF should be plenty of reservoir capacitance. 3. Yey, that looks good!
|
|
|
Post by Peegoo 🏁 on Feb 26, 2023 6:58:09 GMT -5
I feel conflicted for advising about the EZ81 heater issue now. Do you mean running the EZ81 on the common heater rail is not ill advised? I didn't change to silicon rectification because of that; it was more for reliability...so no worries on that. I will add that fuse before the choke and delete C1a. Does leaving R1A (HT bleed-down) in there detract from anything? Some amps have this and some don't. Thank you sir! First three pints and a dram are on me if we ever cross paths. Geno
|
|
|
Post by Peegoo 🏁 on Feb 26, 2023 7:43:06 GMT -5
I got a laff out of Merlin's section header, "Implementing a Standby Switch (The Least Bad Way)."
Funny but true stuff!
That's like the statement, "everything is a fuse if you use it wrong enough."
|
|
|
Post by Peegoo 🏁 on Feb 26, 2023 15:04:10 GMT -5
Okay...going to press with it. Here's the final roadmap to tonal nirvana. [ahem] OR perhaps a bunch of sparks, smoke, and a story the neighbors will tell for many years to come.
|
|
|
Post by Pinetree on Feb 26, 2023 16:32:31 GMT -5
No bright switch?
|
|
pdf64
Wholenote
Posts: 557
|
Post by pdf64 on Feb 26, 2023 18:08:52 GMT -5
Bugger, I thought i sent a response but it seems to have been lost in the ether. I’ll try again tomorrow, but yes, proceed.
|
|
|
Post by Peegoo 🏁 on Feb 26, 2023 19:58:51 GMT -5
There is one. It's in the circuit right before J1, and it's called a Telecaster. If it needs a switch I'll add it. I put one in my previous build and it's a nice feature.
|
|
|
Post by Pinetree on Feb 27, 2023 22:16:27 GMT -5
We're gonna need a lot of photos.
|
|
pdf64
Wholenote
Posts: 557
|
Post by pdf64 on Mar 2, 2023 5:05:56 GMT -5
I feel conflicted for advising about the EZ81 heater issue now. Do you mean running the EZ81 on the common heater rail is not ill advised? I didn't change to silicon rectification because of that; it was more for reliability...so no worries on that. Dunno whether it’s ill advised, I’m sure Merlin wouldn’t put it forward as an option if that was the case. Rather just that it opens the door to another potential failure mode. And most commercial designs use a separate winding for an EZ80/81. Fusing the HT winding would mitigate for it. And I’m a big fan of the slow ramp up of the HT voltage which indirectly heated valve rectifiers provide. I don’t think it causes any problems per se, and it should help to ensure that the HT caps don’t zap you when you tinker inside the chassis. I don’t fit them, as I have a hypothesis that leaving some residual charge in electrolytic caps might help them to last longer. If you haven’t done the heater wiring yet, here’s my little tweak to Merlin’s method (ie rotate the socket and use the gap between 1 and 9 to feed the wire through, leaving plenty of clearance between wire and lugs)
|
|
|
Post by Peegoo 🏁 on Mar 2, 2023 11:48:05 GMT -5
I don’t fit them, as I have a hypothesis that leaving some residual charge in electrolytic caps might help them to last longer. Now that is interesting. It does make sense to me--based on what I know about lithium battery chemistry. I know batteries and caps are two different things, but running a rechargeable battery completely flat usually makes it incapable of receiving a full charge, if it recharges at all. Good tip on routing the heaters. I've taken to running them flat against the chassis, usually in a corner like your pic shows. I've tried it two different ways (this way, and flying them in after all other connections are made) and this one here is usually quieter. Cheers and thank you.
|
|
|
Post by Peegoo 🏁 on Mar 2, 2023 11:48:58 GMT -5
I have not started soldering yet...still gathering the parts.
|
|
|
Post by Pinetree on Mar 2, 2023 12:14:29 GMT -5
It's a shame that you bled all over the chassis.
|
|