 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.193.201.123
In Reply to: RE: Wurlitzer 6240 organ amp help needed.... posted by Eli Duttman on November 29, 2009 at 12:23:53
Thanks Eli,
Do you mean C5 and R14? If so do you think I could just remove?
Bear with me as I am just learning. Mostly a kit builder and parts swapper so far.
I'm not after more power as such but I like your ideas if it will make the 6L6's perform better. As I understand I could go either the 6AN8 (mkIII) route or cascoding the 6SN7 as 2 different options for more gain?
Here's a pic of one of the monoblocks. Notice the size of the P/T. I am not opposed to completely gutting and using a differnt topology. But the OPT would probably have to be replaced and still fit underneath.
-D
Follow Ups:
Provide a picture of the underside, please. I want a look at the O/P "iron", among other things.
The topside photo suggests that the 4 pin connector has been deactivated. I see hints of a RCA female and a power cord. If said power cord is not 3 wire safety grounded, it has to go.
Neff's idea about using the 4 pin real estate for another tube makes sense. However, I'd put a hole shrinker there and install a high gm miniature type.
A Mullard style circuit employing a 6AB4 or 1/2 a 12AT7 as the voltage amplifier and an ECC99 as the LTP will provide ALL necessary gain. SS rectification of the B+ might be necessary to prevent power trafo overheating. An ECC99 draws more heater current than a 6SN7 does and the additional voltage gain tube is yet more strain. While that looks like a fairly hefty hunk of magnetics, wrecking it due to carelessness should not happen.
Eli D.
Yeah, that's what got me going on this project. I finally got a hold of some panel mount IEC connectors. That pic is from when I brought them home. Spent last night and this morning cutting in and installing a proper grounded 3 wire power supply, a power switch, proper binding posts, better RCA'a and a pilot light. That 4 pin connector has been removed and that's where I put the power switch but it can easily be moved if I need to add another tube in that locale.Like I said the previous owner had started the mods but I kind of had to start over. Some of the caps had been replaced but not all. Also a lot of the resistors are really out of spec. and are some really old style. Oh wait, I think you responded to my old post about that type of resistor.
Now that I feel safe about plugging them in I'll start on the resistors and order up the caps.
Thanks for more ideas. What if I added another 6SN7? I thought I've seen a 6L6 AMP using 2 6SN7's but can't find it again.
-Dogwan
Edits: 11/29/09
2X 6SN7s as the small signal complement is the "classic" Williamson style circuit. It works well, IF the O/P trafo is absolutely top notch. Otherwise you get ( YUCK!! ) a phase shift oscillator.
Mullard style circuitry is "classic" too, but it is much less prone to stability trouble than Williamson style is.
Please provide a photo of the underside that includes the O/P transformer.
Eli D.
Here ya go, again this is an old pic before I started work on it.
That O/P trafo does not look particularly hefty. I strongly suspect the tubes installed in the "beast" now are 6L6GCs as the "finals" and a 5U4GB as the rectifier. My "educated guess" is that OEM was (respectively) 6L6s (metal envelope/19 W. max. plate dissipation) and a 5U4G. There's no problem with the later variants being in situ, as they will "loaf".
IIRC, you previously indicated that O/P trafo replacement was a possibility. Edcor's model CXPP60-MS-6.6K seems suitable. The ultra-linear (UL) taps are a blessing. Full pentode mode requires a lot of NFB of 1 kind or another to obtain a satisfactory damping factor. UL mode requires less help from NFB to be acceptable in the damping factor dept.
Eli D.
If I want to add another driver stage and beef up the OPT I think I would pretty much have to scrap the chassis. At which point the only thing worth salvaging (aside from the I/O parts I've added) would be the power tranny.
Might be a fun project. Maybe go with a different output tube at that point.
The learning experience is why I picked these amps up in the 1st place. A scratch build would definitely challenge me.
-Dogwan
Those chassis have a sweet appearance. Remove the can cap & use seperate caps under the chassis. A better audio transformer should be able to fit in the new space. Edcor has some interesting products and many of the audio transformers are only 3.2" wide. Smallest Edcor high quality full bandwidth PP transformer is rated 25-watts and superior to the existing audio transformer.
Before you start a scratch build project, do something constructive with the Wurlitzer monoblocks.
Let's leverage the 30 W plate dissipation max. of the 6L6GC compared to the 19 W. max. of a metal envelope 6L6, while allowing for lack of heft in the OEM O/P trafo. SS rectify the B+, to increase the rail voltage. Triode wire each 6L6GC by connecting g2 to the plate via a 100 Ω Carbon film resistor. A small toroidal trafo will be installed to energize a bias rail (C-) for the "finals". "Fixed" bias will allow for the greatest possible plate to cathode differential. Power O/P will be less than that of the OEM setup, but strain on the O/P "iron" is lowered and loop NFB requirements are reduced.
It will not be the end of the earth if a Mullard circuit's LTP is a 6SN7, as gm is still reasonably high and R P is reasonably low. LTP performance will be enhanced, if the tail resistor is replaced by a 10M45S CCS. Mount a 7 pin mini socket for a 6AB4 voltage amplifier. 200 to 220 V. on the plate and I B = 3 mA. is an excellent set of operating conditions for the voltage amplifier. The 5 VAC winding will be used to energize the 6AB4's heater. Hopefully, regulation effects will bring the actual voltage within 6.3 VAC +/- 5%. It's not inconceivable that a small series resistance will be needed to prevent an overvoltage condition. Remember, a 5U4's filament draws 3 A., while a 6AB4's heater draws 150 mA.
Eli D.
Think one of the multi-cap cans gave out. I had the bottom open so I could double check the power tranny spec's. Heard a crinkling popping sound then the power fuse blew. Replaced the fuse then turned on for just a 1/2 sec and got blue flames in the 5U4GB.
Nice thing about mono-blocks is you can compare readings. Was worried it was the PT so I compared resistance readings across all windings between the 2 PT's and they matched up. But when I compared C8's I couldn't get any resistance or capacitance readings from the suspect one.
Anyway I did find the spec's for the PT online: 380,0,380 @ 150ma, 5v @ 3a, and 6.3v @ 3a.
I was trying to see if these would support a Dynaco MKIII or IV clone. If I move just one of the tube sockets I should be able to squeeze the Edcor OPT in on top.
Take a look at the attached schem. Do you think this could work on the chassis with the existing PT's and those Edcore's?
Thanks for all your help,
-Dogwan
"Take a look at the attached schem. Do you think this could work on the chassis with the existing PT's and those Edcore's?"
No go! :> ( 2X EL34s draw all 3 A. the 6.3 V. winding can supply. 2X 6L6GCs draw 1.8 A. from the filament winding.
3 A. is enough to power 2X 6L6GCs (1.8 A.), an ECC99 (0.8 A.), and a 12AT7 section (0.15 A.). :> ) The total is 2.75 A.
Look at the RCA 6L6GC data sheet. Notice the triode plate curves show upwards of 50 V. as the grid bias for rational "idle" currents. More than ever, "fixed" bias is (IMO) the obvious way to go. SS rectifying the B+ will yield a rail of approx. 490 V. A 50 mA. "idle" current seems sensible, as anode dissipation will be under the 30 W. max. allowed.
While it is possible to derive the bias (C-) rail from the 5 VAC winding, I would not do so. A (SIC) 16 stage voltage multiplier occupies lots of space and is not inexpensive. Just tie the 5 VAC wires off. An AnTek (www.antekinc.com) model AN-0124 toroidal trafo costs $10. Connect the 2X 24 V. secondaries in series and Greinacher ("full wave") voltage double the 48 VAC with 2X 100 PIV Schottky diodes to get the negative bias rail. Notice that you will have the option to run the "finals" at a small "idle" current, should you so choose.
Eli D.
But I would need to see it laid out in a schematic.
Re: the Mullard EL34 circuit I forgot to mention that I knew I would have to add a small cheapo 6.3v "helper" transfo. for the heaters, probably a 4a just for the EL34's.
Found this design that I think could work. What do you think?
-Dogwan
"Found this design that I think could work. What do you think?"
That's an "interesting" hodge/podge. A paraphase splitter driving a differential amp. Unfortunately, for good cause, paraphase splitters are out of favor. So, forget that design too.
It seems that you want a well proven "road map". OK, I've got a real good 1 for you. It's the Harman-Kardon Citation V. "Steal" the Cit. 5's small signal circuitry and use "fixed" bias triode wired "finals" into the O/P trafos currently in situ. SS rectify the B+ and use the "5" VAC winding to energize the heater of the 12BY7 voltage amplifier, in a "6" V. setup. The 6CG7 shown in the schematic is 100% electrically equivalent to the 6SN7 now in place. ;> ) AA sponsor Jim McShane is THE expert on H/K tubed Cit. units. Send him an EMail about the project. Undoubtedly, Jim will have parts you can use at a fair price.
Eli D.
Hey Eli,
Hope I'm not testing your patience too much.
You're right I am leaning towards a proven design I can follow easily. I like the Cit.V idea as well as the 6AN8 idea from the Dynaco mkIII.
So while I was out for a run in the park today I was thinking about this project. With so many options on the table I realized I should list out my goals and prioritze from there.
1. Use existing PT's and chassis, structural modification if necessary.
2. Go UL and upgrade OPT's.
3. Achieve 25-30w output.
4. Cathode bias for simplicity, Or if fixed bias add toroidal xfmr. as per your suggestion. (More power would be nice but I like the more lush tube sound)
5. If possible retain 1 tube for driver/phase splitter (6AN8?) to stay within capacity of PT filament circuit. If needed could add helper small helper xfmr., maybe combine with toroidal for C- rail if fixed bias.
6. Stay within 6L6 family for output stage. 6L6GC, KT66, KT77 ?
7. No preference for SS or tube rectification. Tubes are more romantic, SS could free up some mA's if needed.
I don't want to run this thread on forever but I did find this out there on the internet. It was on a blog site, most of it was in Japanese. To my uneducated eyes it looks similar to the Dynaco driver stage. Of course I would need to make sure it is a viable design. But I'm intrigued.
-Dogwan
Stay away from the KT77 and KT66. The '77 is like an EL34, while the '66 is electrically similar to the 6L6, it draws more heater current.
If you are going to go UL, buy Edcor CXPP100-MS-6.6K trafos. The extra magnetic headroom is insurance against core saturation from the GNFB loop. Also, you keep the option of Class "AB2" operation open, down the road. Big "iron" never hurts.
I don't know where you got 290 Ω for the cathode resistor value. Remember, upwards of 50 bias Volts are needed. 50 V. and an "idle" of 50 mA. works out to 1 KOhm. Can you say heat ? Dude, go combination bias. Build the C- supply. Put a shared 100 Ω/330 μF. network under the cathodes. That's a single grid voltage adjustment and very stable. 10 V. across the 100 Ω resistor = 50 mA. "idle" per tube. :> D
SS rectify the B+ with what I call a "cockeyed" bridge. Construct series wired pairs of UF4007s. Parallel each pair with a 10 nF. high WVDC snubber. Connect the cathode ends of the assemblies to the ends of the rectifier winding. Connect the assemblies' anode ends to ground. Connect the anode of a 1200 PIV Schottky diode to the CT of the rectifier winding. Connect the Schottky's cathode to a CL150 inrush current limiter. Connect the CL150 to the PSU. That setup is QUIET and has a slightly softened start. Use 525 WVDC twistlok (FP) caps. in the PSU filter. Never ignore start up surge, regardless of rectification style.
Finally, install some protection for the screen grids of the "finals". Put 100 Ω resistors in the lines leading to the UL taps.
Eli D.
"I don't know where you got 290 Ω for the cathode resistor value"
- That wasn't my doing. Thanks for the catch. Like I said I mined this from someone else (copied a JPG.) while looking for PP 6L6GC ideas. I would definitely verify all values and layout before building.
"buy Edcor CXPP100-MS-6.6K trafos"
-Do you mean the CXPP60-MS-6.6K? Wouldn't that be more than adequate? Oh wait, as I type this I realized why you recommend.
So I'm liking this 6AN8/6L6GC UL PP direction. I'll start saving up for the the trafos. and acquiring some hardware. Also I'll also draw up the schem's a little better and start verifying values.
I like some of your ideas. Won't be starting right away. Have another project (vintage SS, aaargh!) that I've been putting off. I will re-post and ping you when I get closer to launch.
-Dogwan
I wouldn't remove them. Check the resistance across R14 for out-of-tolerance rating. May want to replace C5 (and R14) anyhow. Be sure to use same wattage and voltage rating on the replacement components (no lower ratings).
What would be the repercussions if I did remove the Zobel network?
I'm tempted to try and see what the sound is like but don't want to blow anything up.
I've looked around at other schematics of similar PP designs and haven't found any that use a Zobel before the OPT. But I know I know only enough to get myself in trouble.
This is not a Zobel network even though it looks like one. It is a resistive load so some load is always on this high output impedance tube. With no load and global NFB the output pentodes can overshoot and possibly oscillate at voltages above their voltage rating. So it is a dummy load for stability at all times and to protect the pentodes for when it's accidentally playing with no speaker plugged in.
Zobel networks would be for flattening average speaker loads combined, to be more resistive across the band to the pentodes. This is a 33K resistor load only, the cap is just a DC blocking cap and not a Zobel forming cap. The -3 dB point on the LF rolloff is down to 9.7 Hz. Not in band for the speaker load at all.
-Kurt
I believe the correct term for an RC network in the primary of an OPT is "conjunctive filter". Seems to be a rather popular guitar amp tweek. Search the net for all of whys and wherefores.
Craig
I think you are right. I made a miscalculation anyway. It will cut down energy in the treble, not all the way to 10 Hz. I have made tricks like this in the input, but I see the advantage here is that when things go a little overload, this network can calm down the treble still for that "smoother" sound. Now I know a guitar amp trick.
-Kurt
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
