• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Poindexter's 6v6 PP amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Yes, Siemens 6922 with the gold pins. I rolled a lot (I mean a huge lot) of different 6DJ8, 6922 & 7308s plus other types. The Siemens was the second best tube. It was ever so slightly 'edged out' by the very rare Raytheon 6CG7 black plate in the red/white/blue box. The Raytheon 6CG7 only provides about 14 volts output with one volt input. It requires a line-stage preamp for full volume.

Raytheon 6CG7 or any Raytheon with a 6CG7/6FQ7 in the white with red color box are made in Japan (stay far away from that tube). The Siemens were at its voltage limit as I used 330 volts on the B+ rail, but the tubes never arc'ed over. At the 'poinz' design of 315 volts B+, no problem. Do understand the 6922 barely provides full power output. At the common bias of -20 volts in the 'poinz' design, the 6922 provide up to 18 volts or so output. A high quality line-stage preamp does provide slightly better dynamics with the 'poinz' amp. You can go this route at a later date.

I also tried many 5965, Amperex pinch waist E180CC (and other E180CC), 12AV7, etc. Forget it, the Siemens 1960s era A0 code with grey shield as viewed at the top mica area were very nice tubes. Also, RCA imported Siemens to the USA during the early 1960s. The RCA boxes are usually stamped in black ink 'Made in germany' & same is on the tube glass. Per my 2 cents, you just can not go wrong with this 'poinz' design tube amp & the 6922.

The new manufacture Electro Harmonic 6V6 or believe it or not GE 6V6GTA 'coin base' tubes are very good.
 
I better add this......

Do not compromise the design with less than better quality parts as mentioned on the 'poinz' site.

My amps are monoblocks with Vishey resistors, AuriCap coupling caps, MQ iron, an LCLC power supply (each amp) using an oil cap on the output. Another LC section feeds the phase-splitter tube with a Solen polyproplyene cap. Electrolytic is not good in that position! The AuriCap requires 20 hours to sound good & over 100 hours to smooth out in the high frequency registers.

Do not use less than a 8k primary audio transformer with 300 to 315 volts plate. 10K would be better if one could be found. I use about 5.6K primary with parallel PP 6V6 tubes.

The rectifier that handily outperform the others I auditioned are the Sylvania 5R4GYB black plate. The tube is the design with the plates spaced 1/8" from each other in the center.

A few of us will state these amps ever so slightly outperformed a PP 45 I built with the same chassis, power supply design, quality nickel core MQ iron (both amps use MQ), a direct coupled 6P5GT to 6SN7, and same rectifier tube. The PP 45 amps outperformed a well known SET 45 on high-efficient quality speakers in the upper & lower audio frequencies per the new owner. A SET does have a pleasing midrange that is almost intoxicating to many audiophiles. The PP45s outperformed the standard amp with the 5965 tube. I did hear some somewhat pleasing midrange warmth with a RCA 5965 triple mica tube.

It may appear to some I am bragging. Not so, just what four audiophiles heard including myself. All of us prefer a neutral sounding amp without any midrange smearing often called warmth. Sonics are subjective & all audiophiles hear differently. The stock 'poinz' 6V6 amps are excellent & an inexpensive design superior to many amps. One design intent is to eliminate an expensive preamp that many simply can not afford & keep parts costs within reason.

If I insult some by sounding like a 'know it all', no flamer intended. I have much to learn in audio.
 
Hey-Hey!!!,
I would suggest hunting for output Iron which offers some use past the amp it was first purchased for. Out of the latest amps I have designed and built, only one will stay together and remain while a second set of Iron is acquired.

Amperex, you are a bit out of my easy travelin' radius. Do try to visit if you're in the neighborhood. I'd love to have you and your gear over for a listen.

I found an easy way to get custom Iron: have it wound to spec. Heyboer is one of my favourite sources. They are the only source of audio Iron which will send the bill with the Iron, and expect the funds within 30 days. They even did this for the first stuff I ordered, fromthe first I telephoned them and worked out the details. Power Iron, signal Iron, and output Iron.

Heyboer is making the custom 10k output for me. 2 pairs of E-Linear taps, one for the front end, and the other at a different percentage for the g2/U-L connection. One spare pair for experimentation while I was at it, for 20%, 30%, and 40%. The basemodel OPTx I used for a model has a versatile winding geometry for applying taps at the end of the layers in 10% increments. AFIK, they are the only filks who are comfortable doing this. Suprisingly, I approached MQ about this and was refused on the tapping issue. Their loss, as that amp works very well, and it is sooo simple.

I do like pentodes for simple amps. Two stages, single B+, short path NFB and Class A bias on the finals. I have a HY69 amp like this, and am building a big brother to it, with the 5X125B. Good ole Thoriated Tungsten filaments. Fixed bias, cascode front end and my E-Linear NFB scheme.

It requires some CCS-es,but depletion-mode MOSFET ones are not soooo hard. Look on Bas Hornman's page for the DN2540N5 battery and non-battery circuits. Watch dissipation ratings on them, w/o heatsinks, stay below a Watt for the TO220 versions. The LND150N3 don't dissipate much heat and are fine with the TO92 case.

I put the name Merlin to this one, and am playing with it on the ART forum. Aprelim schematic is on the linked projects page:

http://audioroundtable.com/GroupBuild/Projects/

On parts, motor-run caps for the PS. Choke input filter, vacuum diode rectifiers. Oil film caps for the coupling, and PP grid chokes of 250/1k Hy inductance. Mills resistors for the two plate loads. L-C filter for the bias supply using Schottky diodes for minimum noise. Good quality 'lytics for its filters.

When you stay simple, there aren't too many parts to inflate the bill. There is of course more to the Merlin amp, but the simple approach is *VERY* nice.
cheers,
Douglas
 
If I plan a trip your way, we will compare systems. It could be quite educational to both of us. Your amp design looks good. I bet it is a great performer.

I have made great progress in modifying my Pro Ac 2.5s to superior performance.

I have a third generation DIY preamp & starting a forth design.

I finished my latest tube type FM tuner & I am calling it good. It rivals some CDs on the best FM station, but most would never believe that 'CD' statement of mine.
 
Are you sure about those 5965's? Here's what I got from the spec sheet:

Miniature twin triode designed for use in high speed digital computers. Each section features a high zero bias plate current, a sharp cutoff characteristic...

(Emphasis Mine)

That doesn't sound (OK, pun intended) good at all. That is more like the description of a BJT, and the opposite of what you'd like for analog amplification. How do they get these to sound good? :confused:
 
The 5965 is one of the many "computer tubes" or "digital tubes" from the dawn of the computer age. Early in the computer age it was discovered that tubes did not like being operated in cutoff for long periods of time. It seemed that the cathodes would loose emission after lond periods of cutoff, therefore the computer triodes often had a special oxide coating to solve this problem. Most of them also were specified for "class A amplifier operation" including the 5965. They often have a low plate resistance, which is usually a plus. Many of these triodes evolved from the 12AU7 tube, including the 5965, and the 5963.

I have used many of the computer triodes in amplifier circuits with good results. I like the 5963 and the 7044 as driver tubes. The often respected 5687 is one of these types.

The 5965 has no direct substitute, however the 12AY7/6072 or the 12AV7 are close.
 
I bought 10 RCA 5965 triple mica for $25. Inexpensive indeed.

Other bargains:

5814- low distortion 12AU7 type. $4.50 NOS

5963- 12AU7 type. $4.00 NOS

5V6- Use a small resistor on a 6.3 volt xfmr winding. $5.00 NOS

6S4- 2-watt output triode with cathode. $3.50 NOS

12V6- $4 NOS

12FQ7- $5.50 NOS

12SN7- $7 NOS

12J5GT- $3.60 NOS.

12X4- (rectifier) $2 NOS. Parallel for 140 ma @ 350 VDC max.
 
Bandersnatch:

Eli: please quit speaking of why the inexpensive valve actually is useful for audio. It won't be so long before it is expensive with such discussion.

Hey, how else is one to learn? When I came here, I explained that I was strictly a solid state guy. This vacuum tube stuff is still pretty new to me (although I'm getting up to speed on it). Still, there are quite a few "sleepers" out there rather than overpriced "boutique" tubes.

In the final analysis, those digital "number" tubes won't be available unless folks out there actually use them.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.