Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: LooseChange on March 27, 2010, 08:54:37 am
-
I have this little amp that was probably a PA amp of some sort.
I am not sure about the output transformer. Will it work with a speaker? What is this?
I'd like to keep like it is and tweak for guitar.
Any other thoughts would be great.
(http://i40.tinypic.com/t7hjr9.jpg)
Thanks!
-
Looking at the schematic, it appears that the transformer and the inputs are all hooked to a plug (hookup) of some sort. Is that the case ?
-
A pure guess is that the 3.5V out is 4 ohm and the 20V out is 8 ohm. I'm surprised there is no 70V out if it's an old "public address". Pull the tubes and disconnect the FB from the transformer. Put a small AC voltage on the primary and measure the voltage ratio, then calculate the impedance ratio. A pair of 7189's are looking for an 8k load.
http://www.mif.pg.gda.pl/homepages/frank/sheets/135/7/7189A.pdf
-Richard
-
Looking at the schematic, it appears that the transformer and the inputs are all hooked to a plug (hookup) of some sort. Is that the case ?
No. Just terminal lugs.
Thanks Mr. Buttery. I never thought that those voltages would be on the output taps. No 70v tap.
-
Executone was intercom systems, mostly under 1 Watt. But of course they added products to get bigger jobs.
This is a simple basic speech/music audio power amplifier.
The 7189s are essentially EL84. Here they are run at around 13W/tube dissipation, expect maybe 14W clean, 18W guitar-rating.
The output tap labels seem clear but imply odd impedances:
14 Watts at 20V is 28 ohms
14 Watts at 3.5V is 0.9 ohms
14 Watts at (20V-3.5V=) 16.5V is 19 ohms
The "0.9 ohm" tap might power any reasonable number of nearby speakers in parallel at around 1 Watt each: a small-facility announcement system. The 20V tap might be for longer runs feeding "25V" transformered speakers at almost full rated power. I don't think the 25V Standard was common in 1965; Executone may have invented it. Some codes don't like 70V systems without hard-wire, 24V is generally tolerated in cheap cable, and Executone may have picked 20V just to be sure.
For your use, the only thing which looks sensible is to load from "3.5V" to "20V" with 16 ohms.
The signal at V2a grid must be about 0.2V. Before that is a standard James bass/treble control, loss about 10:1, so it needs 2V. V1 gives gain about 50, so input sensitivity is about 40mV.
You can just plug your guitar at Input Hot #1 and Input Gnd. It will play.
Snipping R8 NFB resistor will make it a little more hot and snarly, but the hiss will rise. When you have enough gain for lively guitar, you usually want your volume control after the first stage, not out front like this. (An intercom announcement system would not be set as loud as a gitar amp.)
For guitar.... I'd move the Volume over in front of V2a grid, then re-build V1 with a 7-pin pentode with ~~100K plate load. This will mean changing the token R29 100K C17 0.1u decoupling to more like 10K 10uFd. The James control is fine for guitar, tho you will mostly use the top-half of the knobs.
http://i41.tinypic.com/r0qhvo.jpg
-
> For your use, the only thing which looks sensible is to load from "3.5V" to "20V" with 16 ohms.
You mean I should use those taps together direct to the speaker?
Thanks for the great info... I was going to move the volume as suggested. Not sure about changing V1 to a Pentode. I guess if I find a nice 7 pin I'll consider that.
First order of business is to plug in and get some noise out of it.
-
I have a matching pair of old Executones with 6L6's. They were the PA for a large plant with 2 buildings. They were going to toss them, along with 20 Shure Green Bullets. I saved the amps and one of the Bullets. Sadly, 19 of them went to the dump before I knew it was happening! The problem is they're huge with no preamps. I'm kinda drifting towards tweaking them for a stereo system. One of these days!
Dave
-
I have a matching pair of old Executones with 6L6's. They were the PA for a large plant with 2 buildings. They were going to toss them, along with 20 Shure Green Bullets. I saved the amps and one of the Bullets. Sadly, 19 of them went to the dump before I knew it was happening! The problem is they're huge with no preamps. I'm kinda drifting towards tweaking them for a stereo system. One of these days!
Dave
I have one of these my self I have retro fitted mine for guitar I love it bro
-
I pulled the amp back out and made some noise.
What's up with the power supply center tap? (lower right corner of the schematic, top of thread)
It's at -19volts.
I've seen raised heaters but this?
-
What's up with the power supply center tap? (lower right corner of the schematic, top of thread)
It's at -19volts.
That's an old trick to develop a negative voltage without having to actually have a separate diode/filter. It's most commonly seen on conventional full wave rectifiers but will work with bridges too. Rather than connect the PT CT directly to ground, connect it thru a resistor to ground. Current always flows thru the resistor in the direction to cause a negative voltage at the CT (in respect to the ground).
In this amp, R5 is that resistor and C4 is the filter cap.
-
I've used that trick to drop B+ 20 volts or so. Looks like they were using neg 19 for something since it goes to a terminal. PB... push button? Hammond Organs did this too. I can't remember what they used the neg supply for. That old JVC I made into a Stout had the same deal. It was a power amp out of a console.
-
Okay makes sense.
That cap is a 500uf electrolytic. I figure I can eliminate the resistor and the cap if I do a cap job.
No reason to replace that, right?
-
Yes. Just connect the CT directly to ground and throw R5 and C4 away.
-
Yes. Just connect the CT directly to ground and throw R5 and C4 away.
That brings the B+ up to 380v.
I moved the volume pot and the amp sounds good using the OT taps into a 16ohm speaker per PRR's suggestions.
I've got a 6UA6 7 pin pentode I'd like to install. Any suggestions on the wiring?
Thanks!
-
Grab the values from the 1/3 watter. IIRC 220k plate, 470k screen, 1k8 cathode,33k grid stopper, 1M grid leak, 22uF bypass, and .047 screen to cathode.
-
Thanks!
I reduced R9 (cathode resistor on PI) to 1k to see what would happen. Volume was a lot louder. Would messing with this resistor cause some kind of unusual issue in the PI?
-
R9 is actually not part of the PI. The PI is pins 1,2,3. R9 is just a bias resistor for a gain stage just prior to the PI. Reducing it's value increases the gain of that half of the tube. Reducing it's value 'may' overdrive the PI causing more distortion, but that could be a good thing if you like the sound. It's not dangerous to the PI.
-
What's up with the power supply center tap? (lower right corner of the schematic, top of thread)
It's at -19volts.
That's an old trick to develop a negative voltage without having to actually have a separate diode/filter. It's most commonly seen on conventional full wave rectifiers but will work with bridges too. Rather than connect the PT CT directly to ground, connect it thru a resistor to ground. Current always flows thru the resistor in the direction to cause a negative voltage at the CT (in respect to the ground).
In this amp, R5 is that resistor and C4 is the filter cap.
Is the 500uf cap there to keep the -19 volts DC relatively constant?
Without the cap, the voltage would go down (more negative) as the amp's current draw goes up - correct? If that is correct, seems like you could use it as a "sag resistor" without the cap.
If you use a Zener on the PT center tap instead of the resistor/cap, is that a reasonably well regulated supply of negative DC?
Sorry for the re-direction (a.k.a. "Hi-jacking"). This just got me thinking which always is a dangerous thing.
Cheers,
Chip
-
> I reduced R9 (cathode resistor on PI)
Leave that alone. R8, the tube V2a, R17 set-up a DC voltage which, in this amp, is direct-coupled to V2b, the cathodyne. As designed, V2a plate is at +100V, so is V2b grid, and V2b cathode a few volts higher.
With R8 at 1K, I'd expect plate to fall to +70V, also cathodyne voltages to shift significantly. Since 7189 is a very sensitive tube, it will still work OK, but this probably isn't what you want to do.
> a lot louder
R8 and R9 100K make the NFB voltage divider. As designed, 100K/2K= gain about 50. Your mod, 100K/1K= gain around 100. But you could also leave R8 at 2.2K, and raise R9 to maybe 220K. That also gives gain around 100 but does not change the DC biases around the driver section. You could go still higher, or snip R9 completely, for a more-raw less-PA sound.
If R9 is 100K you may need C5 33pFd. If R9 is higher, C5 doesn't do anything good or bad, leave it or omit it as convenient.
> load from "3.5V" to "20V" with 16 ohms
I should mention: this way, you can NOT use a grounded speaker jack. No danger, cuz there's only 3.5V max on the plug shell, and this amp will not burn-up if the "3.5V" tap does get grounded, it just won't play right. But it may be convenient to use a good metal jack in the chassis.
Disconnect the OT Black wire from ground. Strap the OT Yellow wire to ground. Now the former "3.5V" tap can be called "0V" and grounded; the former "20V" tap may be called 16.5V (or 17 watts in 16 ohms) hot/tip.
> values from the 1/3 watter
The James tonestack is a heavy load; also with V2a in the path we don't need MAXimum gain. I like this condition plagiarized from G.E.:
-
Thanks PRR!
Changed the OT secondaries as suggested.
Added the Pentode (6AU6).
Did not mess with the NFB yet (I think you got R8 and R9 reversed, But I get it).
The amp is sounding pretty good.
I think I need to mess with the tone circuit a bit. Any suggestions to changing the values to get closer to guitar frequencies?
-
Any suggestions to changing the values to get closer to guitar frequencies?
Maybe take a look at some of Tubenit's stuff in the schemo section.....he uses the James & has tweaked it taste like a guitar. :tongue3:
-
I took some voltage readings on the 6AU6 and they seem unusual.
Plate: 28v
Screen: 55v
Cathode: 1v3
The B+ on the other side of the plate resistor: 243v.
I changeed the voltage supply as per PRR with the 22K and the decoupling cap (10uf). The supply up to the 22K is 308.
So, is everything in order?
-
>Any suggestions to changing the values to get closer to guitar frequencies?
My 1st suggestion is to get away from the James stack. I've never found it to be "ideal" for a geetar amp, but assuming you're going minimalist here, I played around with Duncan's Tone Stack Calculator and attached 2 screen captures with controls @ 75%. Note the existing tone stack is bass heavy. That was hifi in the 50's to compensate for less than efficient speakers. I brightened the tone stack up changing 3 resistors and 3 caps (if you consider 470p = to 500p). Worth a shot IMO.
-
> Plate: 28v
> Screen: 55v
> Cathode: 1v3
28V on plate is not enough for strong guitar and high gain. Did G.E. screw-up? Is your tube bad?
Take the cathode resistor up to 1K or 1.5K, we want over 100V at the plate.
I figured the James was there, people use James, not an urgent tweak. Turn the Treb ALL the way up for most bright work, and go easy on the Bass. But it is a wonky stack even for "flat" work like hifi/PA, and for guitar where some specific and steep shapes are useful other stacks have been more popular. The classic Fender BMT is the go-to, has similar impedance and average loss. Alternatively, go back to triode preamp and run a simple one-knob like early DeLuxe.
-
I think the tube is good...
I stuck another one in with the same results. Shouldn't the screen have a lower voltage?
Tried some 6BA6 tubes too. Same results.
You didn't expect that either?
> Take the cathode resistor up to 1K or 1.5K, we want over 100V at the plate.
Will do.
-
Used a 1k5 cathode resistor. Plates are up to 100v and the screen is at 133v.
I ask again, is this okay or should I increase the screen resistor to bring it down.
Resistor values: Plate: 100k, Screen: 220k, Cathode: 1k5.
The amp does sound good. I need to mess with the tone stack.
Bypassing the tone stack provides too much gain. I may go to the classic Fender BMT. (I think that's a sandwich :smiley:)
Thanks!
-
I like that sammich hold the M on a Princeton bun.
-
> I ask again,
For maximum power output, screen voltage high.
For maximum voltage gain, screen voltage low.
In a preamp it may be hard to know "power" from "gain". If it has high unloaded gain but a heavy load (tonestack, mixer, etc) it will have low gain.
For 6AU6 as clean guitar preamp, G2 could be as low as 40V. However that might be too much gain in one stage. And you have another stage of gain (unlike a pentode-preamp Champ).
I don't have a problem with 130V or so.
There is no problem with the plate sitting lower than screen. Oh, plate at 20% of screen gets wonked, but you would not want it that low because you need voltage swing.
You could try 330K 470K on screen. But that will skew your plate voltage. If too far off center, output is lower, you want to change cathode resistor to get plate back to 1/3rd-2/3rd of supply voltage. AFAIK, this all gets too confusing even for paid designers. If it is working fine at 100K 220K 1.5K, sign the blueprint and start production.
> hold the M
Yes, a 2-knob version uses a fixed "M pot" of 5K to 15K. If there is a small hole handy, switch-open the "M resistor" for "too much gain".
-
Blue print signed. I like it just the way it is....
One pentode preamp, A james stack and a simple power amp on the side.
Stay tuned for dessert. A few pictures will follow.
-
The amp is done (for now).
Here is a picture of the amp... It's tiny but powerful.
I used the 6AU6 pentode. The second stage and PI is a 12AX7 and power tubes are two 7189's