Newcomb 6GW8 phono amp- future in the balance

[Frankenamp]

 Hi Guys, here's an interesting critter:
It's an old Newcomb amp chassis that has a 6GW8 (single) tube with a couple of diodes as a rectifier. The original configuration had a little piece of wire that actuated a switch that resembled the backside of a potentiometer/on/off combo pot. That's getting replaced by a regular toggle switch. The resistors that bypass the diodes show 8 & 12 meg on my meter. the .022u /600v cap looks like wax/paper and will be trashed. the (giant) .22u /600V cap is one of those light green Cornell Dubilier plasticky molded jobs, is it ok or not? It has a greyish stripe around one end but I seem to remember that that may or may not be an indication of an electrolytic.  :icon_scratch: The (89T)-> is the tone arm pickup (shielded tinsel wire) that goes to the volume pot (1 meg) The tone control looks strange to me. How much trouble would it be to tweak it to champish (guitar) miniamp duty or should I fugeddabout it and make a cute little ipod squeaker? the inside of the chassis is a bit cramped- 6.25 X 4.25 X 1.25 inches approx... Opinions welcome

[PRR]

The 6GW8 seems to be run at 150% of rating. 45mA and 310V is 14 Watts dissipation in a 9 Watt tube. OK, ~~35mA plate current, "only" 11 Watts. Pretty aggressive.

The old low-volt rectifiers and overbaked equalizing resistors may be replaced with one 1N4007.

How does it sound on iPod? (Use two 1K resistors to Y the two channels together.)

The 0.001uFd across the needle is quite odd. If you are not going to keep the needle, cut that out.

Check the grid voltages. After a minute they should be "zero", <0.1V on triode and <0.5V on power pentode. If power pentode grid is significantly positive, replace the grid cap.

As a gitar amp: input sensitivity is 200mV. You want 50mV to 20mV. You want a gain of 10. System dynamics suggest putting this in front of the present volume control. A booster-pedal is the quick and easy way to try.

The tone control does several things. At one extreme it clobbers the highs above ~~1KHz to kill record-scratch. At the other extreme it enables a "loudness network" which boosts bass (actually cuts midrange) at low Volume-pot settings to give a more-natural reproduction at low listening levels (where you don't hear bass so well). It may be good for a few things. My opinion is that while this was good for a record player, even an iPod, for gitar you probably want to toss it and do something else.

[eleventeen]

Betcha you could rip out essentially everything including the can cap, replacing it with individual electrolytics, and convert its space to a 6V6 and the 6GW8 to a 12AX7 to make an el Champo. The only consideration would be whether the fiament winding could handle the add'l current.

[Frankenamp]

Hmmm interesting ideas, I guess the 6GW8 can take it... Newcomb was known for pretty stodgy if not bulletproof products. I remember a number of them in classrooms a few decades removed and this one seems to have survived that. Ripping out the cap can is one idea, but laying it out would be interesting, given the cramped area- would a 6a/bq5 (smaller bottle 6V6) + 12a_7/ 12BZ7be do-able? the 6V winding uses a #47 lamp parallel with the heater. Remember, this in a 6 by 4 by 1 chassis.  (That's assuming I try another idear above my pay grade.) Couple of other questions about the schematic. I realize that these designers were a lot more subtle than I give them credit- not knowing enough to know why things are done one way or another other than low part count and multiple functions. such as the big honkin .22 uF across the PT's  primary after the fuse, and the smaller .022 uF that goes from the backside of the diodes in the PS to the grid? I did get a little excited, and drill some holes in a piece of steel for a face plate of sorts...

[PRR]

> fiament winding could handle the add'l current

0.050 Amps? I bet it won't mind.

> convert its space to a 6V6

I have a strong suspicion this WAS a 6V6 chassis, but the 6GW8 was cheaper than a 6V6 and a 6AV6.

> rip out essentially everything

Sure. But some phono amps have good karma. I was in the antique store back-room sniffing a few yesterday.

The triode-pentode is a fine tube and there's a niche for a 3-Watt amp.

Depends on condition. "Most of an amp is there", but the can-cap is dubious and the grid caps must be checked. It's work to replace those; but not as much work as replacing everything. Even if you do replace 95% of the little parts, replacing one-by-one is less brain-pain than rip-out and start from blank canvas.

I just noticed the combined B+ and heater winding. You can't use any full-wave (2-diode or FWB) rectifier, and you can't really balance the heater leads. The B+ is OK at this power level because a good fat first-cap is not too expensive, and cleaning the Screen node helps (you will still want the NFB loop and a poor-deep-bass speaker). The unbalanced heater may be OK at the 200mV signal level here, may not be so OK at 20mV guitar sensitivity (though amps with nearly modern sensitivity were sold with unbalanced heaters).

OTOH: You "could" just voltage-double the heater winding for +/-8V and run a 3-watt chip-amp...... I do favor the idea of "least-change".

Also "...the inside of the chassis is a bit cramped..." "Remember, this in a 6 by 4 by 1 chassis"

> a little piece of wire that actuated a switch that resembled the backside of a potentiometer/on/off

The phono had rubber-wheel drive. The wheel would flat-spot if left in-gear with the motor off. You were told to put it in "N"; to enforce this you could not turn off the heat and hum from the amp except by putting the mechanism in N. That pot-back-switch was the cheapest switch available (EVERY radio had one) and you could get the raw switch without a pot. When you put the turntable in 33 or 45, a lever dropped the rubber-wheel on the motor pulley and pulled the string to the switch nub.

[eleventeen]

My reading of spec sheets says a 6BQ5 uses slightly more heater current than a 6V6! Is that possible? (.76 A vs .6 A) If that's something that concerns you. I would favor the 6V6 because the hole in the chassis for the can cap might, if the gods of tube audio are smiling upon you, be the same size as an octal socket!

Newcomb was, I believe, absorbed into a company called Audiotronics, where I worked in North Hollywood, Los Angeles 1982-1983. You're right, they specialized in record players that could survive being dropped down a flight of stairs. That was where I became familiar with the ballistics of improperly installed electrolytic capacitors. When Consuela and Lupe weren't paying proper attention on the assembly line, they would install some number of the electrolytics backwards. They were also very good at diode backwardization. The foil roof insulation above the two older gals who were the very first to plug in newly built monitors (I graduated from record players to the CRT section) was peppered with 1" holes from caps blowing up and thru the foil.

And oddly, even though the chassis is cramped, an octal socket might be a tad easier to deal with than a 9-pin. Depends, I guess, on your eyesight and hand-eye coordination. Also, a 6V6 gives you those 1-2 tie points for the unused pins on the tube, and THAT is useful. In the Amperex tube manual, it appears ALL pins of a 6BQ5 are "internally connected".

Yes, I noticed the oddish combo filament-HV winding on the PT, when I was looking to see how the power supply was grounded. You don't think you could dual-100 ohm that, PRR?

[PRR]

> the big honkin .22 uF across the PT's primary

RF reduction, but it does seem excessive.....

> the smaller .022 uF that goes from the backside of the diodes in the PS to the grid?

I missed that. SCREEN grid. Yes, that's an odd and subtle detail. The evident intent is to inject some 60Hz INto the amp. Why? Because the first filter cap is not large enough to remove "all" the 60Hz from the OT+plate feed. If the cap values are really as-marked, then 0.02u/10U= 1/500th of 250V AC or 0.5V AC is injected at the screen grid. There is about 10VAC ripple at the OT+plate feed, most of that will appear across the OT. It is reasonable that this power-pentode has gain of 10V/0.5V= 20 from screen to plate. I am not sure the phase is correct to cancel....

This is not a stadium blaster. And also it seems to abuse the power section. And the aging filter-caps are dubious. If you re-cap, I'd move the OT feed to the second B+ cap. Half-wave rectifier, 40uFd, 1K 5W, 40uFd will give quite low ripple at the plate and tolerably low hum in the speaker (without cheap-tricks). > 6BQ5 uses slightly more heater current than a 6V6! Is that possible?

No, MUCH more. 6V6=0.45A. EL84/6BQ5=0.76A.

Unavoidable. The 6V6 is an exceptionally efficient tube, designed when PTs cost money and US makers were trying to sell to everybody in the US. The EL84/6BQ5 comes from a different set of thinking, where cost-of-power was not so dear and High Gain was a big virtue. How can a EL84 have so much more transconductance than a 6V6? Only by having a bigger cathode. It does not run any cooler than a 6V6 cathode, so it must eat more heat to stay hot.

The "little" power tubes used in transformerless radios were real heater-pigs. Here. heater power was nearly free, and extra-fat cathodes are the only way to get the Power Output when the B+ is only 100VDC.

[bigsbybender]

How does it sound on iPod? (Use two 1K resistors to Y the two channels together.)

What's the advantage of the resistors in a "Y" versus a play "Y" cable?


j.

[eleventeen]

It's kinda considered bad form to take two audio outputs and just twist them together...implying no isolation between the two separate outputs. Will you get away with it 98 times out of 100? Sure. The times you don't you busted something.

[Frankenamp]

Well, I had this little critter on the back burner and did some poking and prodding around in the little beast. The diodes seemed OK still (at least to my meter) I left the pilot lamp alone and most of the tone stack too. The little .001 uf ceramic cap seems to buried underneath everything else at the bottom of the chassis. (Is this OK for experimental purposes- what does it do?) I just stripped a little tinsel back from the phono lead and connected what was left to the back of a standard input jack via 1M/49K resistors for RFI suppression. If the gain is too low, I assume I can rip out the tone stack and feed straight to the volume control? I'll get busy with a light bulb limiter tomorrow when I'm less tired and try not to duplicate Miss Consuela's trick of sending the old caps through the roof! I guess if it sounds too terrible, I can use the iron to recreate a Champ 5E1/5F1 ish thing with an octal in the cap can position, and a 12ax7 where the old 6GW8 was... I'll have to play around a bit. (Wife got me a new ipod nano to replace the one that went missing; hers went AWOL too- as soon as I opened the box and started looking for the old charger to stick it on the old ipod appeared! Wife found her old one in another similar event. Murphy's law paragraph two subheading A: ...


On a related note my AA764 w/ 12ax7 & 6GW8 might be resurrected- It had no volume & I had no clue... untill an old HAM hand looked at it and noticed that the volume & tone pot leads were too close to the edge of the chassis <insert self-depricating 'moron' icon here> everything else look very nice.

[PRR]

Read back over the thread.

The tone control, when centered, has very little effect. No point trying to mod it out for "more gain".

The amp just doesn't have a lot of gain. Record levels are high and uniform. Pickup levels range all over the place, and more gain is needed for most e-guitar technique.

You can snip the 22K feedback from speaker to driver, but that's not doing much.

Compare/contrast with known-good geetar amps. No respectable amp has just one triode gain stage before a power tube. Triode Champ has two. EL84-like power tube has a bit more gain than 6V6, but like X2, not so you can omit an X50 stage.

Put a LPB-like booster in front. If still lame, WTH, chisel the chassis and build an AA-Champ or some such thing.

[Frankenamp]

Well, the amp does remarkably well so far! I did read over the thread and that's why the questions. I know you have suggested the AA764 champ, but if space is a premium (volume, tone, one jack, and a toggle switch fill out the front fairly quickly). I was thinking more Harvard 6G10, or early Princeton SE to make two staged of gain (or a pentode front end). Just to keep the parts count low, and stay within the confines of a 6.25 X 4.25 X 1.25 inches approx chassis. The P90's have just enough output to make a little fuzz around the edges when everything is dimed. It does have a nice clean sound, as is. and I'm reluctant to fool too much with it untill I get the other full blown champ-ized 6GW8 amp un-Franked and crankin'. (It's a bird in the hand kinda thing) AND, the humor was not lost on me when I was poking my meter leads in the guts of the FrankenChamp wondering why the Red Rocker in my ipod wasn't having the desired effect as a signal source... when the next song to shuffle up was Amy Lee's "Broken"... oh my...