Russian tubes for first Breadboard Project....

[Jack_Hester]

The Breadboard is still under construction, and has slowed considerably for now.  I had today off from the outage, and decided to rest.  However, I've been thinking about what I want as a simple project for a trial run on the Breadboard.  I started browsing my drawings and dug out an older one. 

A few years back when I decided to give tubes another try, I picked up a small kit from Ken Watts for an Epiphone Valve Junior chassis that I had.  I built it, and had no luck getting it to work.  I requested a schematic from him and explained my dilemma, but never got a reply.  I had purchased a number of parts from him, and this was disappointing.  I've since purchased components from him, and have always had good service. 

Anyway, I put it aside for a year or so, and got a large number of schematics, and much experience behind me, before tackling it again.  I contacted him for a schematic, and still got no answer.  So, I drew one from my layout.  That's when I found a missing signal ground in the layout and my build.  I put an alligator jumper in place and it was a really nice sounding little amp.  I downloaded his layout again, and saw a fix in place.  Apparently a problem for others, as well.  As that was the only issue I've ever had with his service, I would say that he is still a good source. 

So, back to my drawings.  I found the schematic that I made of his layout, and decided that it would be something simple to try.  But, with some Russian tubes that I picked up cheap, a few years ago as well.  They are the 6N2P and the 6P14P.  I figured that they would also be a good test of this simple circuit on the Breadboard.  I cleaned up my original and redrew it with these tubes.

I've seen where the 6N2P has been used with satisfactory results.  Has anyone used the 6P14P with the same?  Once I've proven these tubes for myself, I plan to expand on the circuit in a couple directions.  None of that thought out, as of yet.  Today was just a day of resting and drawing.  And, hoping that in the upcoming week, I'll find time to mount the PT, auto-transformer, and some binding posts, on the Breadboard panel. 

Give the circuit a once over, and comment please.  Thanks.  Have a good one. 

Jack

Note: See Reply #10 for the original drawing, combined with a new concept.

[HotBluePlates]

... the 6N2P and the 6P14P. ... I've seen where the 6N2P has been used with satisfactory results.  ...

The 6N2P, from the data sheets I've looked at, is basically a 12AX7 with higher Gm and lower internal resistance (and rewiring for pins 4, 5, 9). Dropped into a 12AX7 socket, it should give slightly more gain.

...  Has anyone used the 6P14P with the same?  ...

If you've bought a new production "EL84" (meaning not expensive U.S. or old western European tubes), then you've used the 6P14P. It's the Russian version of the EL84, and Russian data sheets provide essentially the exact numbers that western data sheets show. So this is the tube that gets imported, marked as an EL84 and sold to you. So it shouldn't be a problem.

Give the circuit a once over, and comment please. 

I'd move R1 (10kΩ) to a point between R2 (1MΩ) and V1 pin 2, perhaps mounted directly on pin 2. I'd also bump that resistor up to 33-68kΩ, as 10kΩ is probably too small to perform much RF reduction, if it was needed.

R5 (91kΩ) seems abnormally small for a resistor to ground right after a gain stage. The Valve Jr uses a 1MΩ in this spot (and I probably wouldn't use one at all). The 91kΩ will severely load the previous gain stage and cut amplification (and maybe introduce distortion; I'd have to plot some loadlines to be certain).

R12 (240Ω) may be a bit high; the stock value is 220Ω. Best determined on your breadboard. I'd keep resistors able to make up 150-270Ω total, as needed, though EL84's tend to bias up with smaller cathode resistors than 6L6/6V6.

R11 (1kΩ) may be a bit high. The stock amp doesn't use a screen resistor at all, and other EL84 amps typically use lowish values here (100Ω), if at all.
   - The reason for that is the B+ in EL84 amps is typically low, and if you make a screen resistor large, it will cut a lot of voltage when screen current goes up. If screen voltage drops with signal, plate current and therefore power output drops. If you use a "too large value" then you can get compression when playing, similar to or more than a tube rectifier. So experiment with screen resistor value from 0Ω (most power output, fastest response; no screen protection) to as high as you like (restricted power output, compression with strong signals; screen protected from over-dissipation).

C5 (1000uF) seems excessively large, but shouldn't hurt. If it were removed, the only change should be a somewhat stronger input needed for the same output power (maybe you turn up to 7 instead of 5 for onset of distortion); the amount of signal increase required will depend on your choice of R12 (bigger resistor, more local feedback, more signal input needed).

C10 (1uF) seems unnecessary. If you're looking to bypass a power supply electrolytic, you should use a value no bigger than 1/100th the existing cap. Further, the point of bypassing is to use a technically superior cap to overcome the shortcomings of an electrolytic. Therefore, this should be a film cap, probably polypropylene (or some more exotic dielectric).
   - I'm thinking the caps with the heaviest current demands, or largest values, would benefit most from bypassing. So I'd look to maybe C7; but I don't have much experience actually using a bypass of this sort.

R13 (220kΩ) has no power rating. I'd probably use 470kΩ or so here, just to reduce power dissipation in the resistor. I don't know under what circumstances you might have the voltages in parentheses, but assuming worst-case of those, P = 400v²/220kΩ = ~0.7w, so make it a 2w. Or ~0.3w for a 470kΩ, and 1w is sufficient.

[Jack_Hester]

HBP -

Thanks ever so much for the suggestions.  The drawing (as shown) is taken directly from the layout.  Now, you've given me some really good tweaks to try.  I'll build, as drawn, and then implement all your suggestions.  I'll go ahead and make a revision to my drawing, to reflect your suggested changes.  Then, I'll post it, as well.  Thanks again for the in-depth critique.  Have a good one. 

Jack

[HotBluePlates]

...  The drawing (as shown) is taken directly from the layout.  ...

I think I found the layout you're talking about (the "Marshallized layout"). In that case, your drawing matches the layout. I suppose it's worth at least trying it their way, just to see what you think about any differences.

If you look at the loadline attached, you'll see what I mean about the 91kΩ reducing gain. I have plotted a 100kΩ plate load (red line) and a 1.5kΩ cathode resistor (blue line) to find the d.c. condition for the stage. It idles a hair below 0.75mA, or 1.125v across a 1.5kΩ resistor.

The coupling cap places the 91kΩ to ground and the 1MΩ volume pot in parallel with the 100kΩ plate load, as far as a.c. is concerned. 91kΩ ll 1MΩ = 83.4kΩ. 100kΩ ll 83.4kΩ = ~45.5kΩ, which is represented by the green line on the curves. This a.c. loadline passes through the d.c. operating point.

The vertical purple line is the idle point, and shows the plate voltage is ~121v with a grid voltage of -1v. I've moved this idle point a little to the -1v gridline because it means a little less guessing on positioning of the other grid voltage points. The thick orange vertical lines represent a shift of grid voltage +/-1v. I estimate the shorter orange line to be at ~150v (maybe a hair more), and the longer line at ~78v.

Gain is change of plate voltage divided by change of grid voltage, so Gain = (150-78)/(-2-0) = -36.

The skinny orange vertical lines show where the same grid voltages would land on the 100kΩ loadline (whose a.c. loadline would only be reduced to ~91kΩ with only the 1MΩ volume pot following the coupling cap). You can see the position of these lines is beyond either side of the thick orange lines, which indicates greater plate voltage output (and more gain). I estimated the position of those lines to show a gain of (172-57)/2 = 57.5.

The interesting thing to note is the distance along the green 43.5kΩ line from a grid voltage of -1v to a voltage of 0v is very much longer than the distance from a grid voltage of -1v to -2v. For zero distortion, they should be exactly equal. They are not exactly equal distances along the 100kΩ loadline, but they are not as-different as on the 43.5kΩ line. In fact, if you had a 1v peak input signal (you never will with a guitar), the 100kΩ line has 6.5% 2nd harmonic distortion, while the 43.5kΩ line has 11.1% 2nd harmonic distortion.

The distortion will be less with smaller inputs (average guitar input will only be ~1/10th that 1v peak case), but shows how the low resistance load alone has caused the gain stage to have essentially doubled distortion. This is also an interesting example for folks who get stuck with pedal and amp-knob naming conventions and think gain = distortion.

[Jack_Hester]

I think I found the layout you're talking about (the "Marshallized layout"). In that case, your drawing matches the layout. I suppose it's worth at least trying it their way, just to see what you think about any differences.

You found the correct layout.  Trying it 'as is' was just to get it on the board, and functional.  And, to iron out and change any breadboard issues.  Then, try your changes. 

The original build had plenty gain, as I remember having to dial back on the guitar volume, to keep it clean with the amp volume turned up.  He has a simple gain mod, to go along with it.  I can't remember exactly what it looked like, but I'm thinking it was a means just to attenuate the signal. 

For now, giving these tubes a trial is one of my goals.  I really need to organize my tubes a bit better.  I've got all my standard tubes segregated into storage boxes with labels.  I've got all my Russian tubes in the boxes that they came in, with the outside of the boxes marked.  I need to put them in the same type transparent storage boxes that the others are in. 

And then, theres piles of tubes that have been given to me, or I picked from various flea markets and such.  None of these have been documented, and I know that I have some goodies in those.  I just haven't made myself sit down and sort. 

Anyway, I'll continue on with the breadboard construction, and get your recommendations on  another revision of the drawing. 

Thanks again, for your many comments.

Jack

[Jack_Hester]

What about the 6P3S-E's?  Got several of those that I can try.  I have been unable to find a clear data sheet on these.  As near as I can tell, they may be a likely substitute for the EL34 or 6CA7. 

Jack

[tubeswell]

What about the 6P3S-E's?  Got several of those that I can try.  I have been unable to find a clear data sheet on these. ..
Jack

General characteristics:

Filament voltage, Volt: 6.3
Anode voltage, Volt: 250
1st grid voltage, Volt: -14
2nd grid voltage, Volt: 250
Filament (heater) current, mA: 880±40
Anode current, mA: 73±13
2nd grid current, mA: no more than 6
1st grid reverse current, mkA: 0.5
Cathode current, mA: -
Output power, W: no less than 5.8
Mutual conductance, mA/V: 5.2 to 6.8
Internal resistance, kOm: no more than 65
Inter electrode capacitance, pF::
- input: 11
- output: 6.7
- transfer: no more than 1
Operation time, h: over 5000

Limited operating values:

Filament voltage, V: 6-6.6
Anode voltage, V: 250
2nd grid voltage, V: 250
Cathode - heater voltage, V: 90 / -200
Cathode current, mA: 90
Anode dissipation, W: 20.5
2nd grid dissipation, W: 2
Resistance in 1st grid circuit, kOm: 150

Operating environmental conditions :

Acceleration of vibration loads, g: 3
- by frequencies, Hz: 5 to 300
Acceleration of multiple impacts, g: 12
Acceleration of single impact, g: 100
Continuos acceleration, g: 100
Ambient temperature, °C: -60 to +160
Relative humidity at up to 40°C, %: 98

http://www.tubes.ru/techinfo/HiFiAudio/6p3s.html

http://www.glowbugs.info/2010/02/6p3-beam-power-tetrode.html

[HotBluePlates]

What about the 6P3S-E's?  ... I have been unable to find a clear data sheet on these.  ...

I recommend Frank's for data sheets.

The 6P3S-E, being a Russian tube, has a data sheet in Russian. Tubeswell has posted a translation.

As near as I can tell, they may be a likely substitute for the EL34 or 6CA7. 

No, they are a good substitute for the 6L6G. Not 6L6GB or 6L6GC. This is the tube that gave Russian tubes a bad name.

Look at plate dissipation first: 20.5w, which is comparable to the metal 6L6 and glass 6L6G rating of 19w.
Next look at transconductance/mutual conductance: 6 mA/V (Tubeswell gave to high & low acceptable limits)

A 6L6 has a transconductance of 6000 micromhos according to the RCA 6L6 data sheet. In fact, much of the Russian data sheet has the exact same numbers as corresponding parts from the 6L6 data sheet. "1000 micromhos" (a unit equivalent to 1/Ω) can be written as "1 mA/V" or "1 milliamp of plate current change per 1v of grid voltage change."

An EL34 has a mutual conductance more like 11mA/V, which means it takes only about half as much drive voltage to develop the same plate current (and output power). So the 6P3SE would be much less sensitive than an EL34.

These tubes (6P3S & 6P3SE) were sold in the U.S. marked as "5881" and "6L6GC", and tended to die in later American amps which pushed the 6L6 beyond its early ratings (and really needed a 6L6GC). In reality, these tubes aren't bad if you operate them as-designed. See the pics below for 6P3SE (often sold as the "coin base 5881WXT") and 6P3S.

[Jack_Hester]

No, they are a good substitute for the 6L6G. Not 6L6GB or 6L6GC. This is the tube that gave Russian tubes a bad name.

I've got quite a few of the metal-cased 6L6's, which I assumed to be 6L6G's.  So, it looks like I've got more of the same, with these.  Oh well.  At least I can see them glow.  Metal-cased tubes contribute little to the ambiance associated with a fine tube amp. 

They won't be on the breadboard, first.  But, they should work well on upcoming experiments. 

Jack

[HotBluePlates]

I've got quite a few of the metal-cased 6L6's, which I assumed to be 6L6G's. 

Good correction!

The 6L6 was introduced in metal form first, then the 6L6G (for "glass") came along, but with the same ratings. I believe the 6L6GA is the same tube, but you never seem to see the "A" version without some other modifying suffixes (like 6L6GAY, for the low-loss micanol base version; go ahead and get your chuckles out...).

[Jack_Hester]

Ok.  How about some 6P6S tubes?  Got a handful of those.  And, it just so happens that the source that I modeled my first drawing by has a new offering that I'm hoping will be suitable for this tube.  Look for it on his home page.

And amazingly enough, it much like the first.  The new one has two inputs with 68K grid stoppers (R1a, R1b), and no 91K (R5) at the junction of the coupling cap (C2) and the Volume pot (R6).  There are a few more subtle differences, but it's basically the same with a different power tube. 

HBP -

Maybe by this evening, I'll have a drawing, implementing your suggestions.

For now, I've combined both drawings in one PDF.  Take a look and comment, please. 

Jack

[eleventeen]

My experience is that those early straight-sided "6L6" tubes with the pin wires coming thru the "squeezed" glass structure at the bottom of the tube (vs the more modern individual glass-to-metal seals that virtually all modern-made power tubes have) will die comparatively rapid and generally catastrophic deaths in Fender amps. They don't just lose emission and get weak, they short out or maybe the internal connection to the cathode opens up. I've had ones where the heater lights up and there is not one freaking electron of output, none, zero.

[HotBluePlates]

How about some 6P6S tubes? ... new offering that I'm hoping will be suitable for this tube.  ...

Again, see the 6P6S data sheet at Frank's. That site has a search function for you to find data sheets.

Within its maximum ratings, this tube is equivalent to a 6V6. But note the maximum screen and plate voltages: 310v and 350v, respectively.

Yes, American amp makers ran riot all over maximum ratings for tubes, possibly with a nod from the tube manufacturers. You're not gonna get the same thumb's-up from Russian manufacturers. I presume that if you respect these tubes' max ratings, you'll get just as good of service life as with American tubes.

My experience is that those early straight-sided "6L6" tubes with the pin wires coming thru the "squeezed" glass structure at the bottom of the tube (vs the more modern individual glass-to-metal seals that virtually all modern-made power tubes have) will die comparatively rapid and generally catastrophic deaths in Fender amps.

That has not been my experience with those tubes. I've never had one die.

But again: these particular tubes are rated 400v plate and 300v screen maximum, and only 20.5w max dissipation. Maybe we get used to cheating American tubes well above voltage ratings, while the Russian tubes are spec'd close to their true breaking point.

For now, I've combined both drawings in one PDF.  Take a look and comment, please.

I don't see any errors.

[Jack_Hester]

That's going to be the beauty of having an built-in auto-transformer on the breadboard.  I can tweak the feed voltage and control the overall B+ to maintain safe operating voltages for these tubes.  Filament supplies will be separate and fixed.  I plan to have separate thermistors in line with each of the transformer primaries. 

Jack

I meant to say thermistor and not MOV, in my statement above.  I've been working with MOV's at the plant, this outage, and had it on the brain.

[Jack_Hester]

I don't see any errors.

I appreciate you verifying for me.  I proof-read, but still seem to overlook the obvious at times. 

I'm tempted to dig out the original and make some accurate voltage readings, as I can't remember why I have dual voltages indicated.  I'll have to find it, in amongst the 'conquered' toys that my wife says of the piles we 'boys' make of them, when we are finished.

Jack

[tubeswell]

My experience is that those early straight-sided "6L6" tubes with the pin wires coming thru the "squeezed" glass structure at the bottom of the tube (vs the more modern individual glass-to-metal seals that virtually all modern-made power tubes have) will die comparatively rapid and generally catastrophic deaths in Fender amps.

That has not been my experience with those tubes. I've never had one die.

But again: these particular tubes are rated 400v plate and 300v screen maximum, and only 20.5w max dissipation. Maybe we get used to cheating American tubes well above voltage ratings, while the Russian tubes are spec'd close to their true breaking point.

I've run the 6P3Se at 460V (biased at 70% of 20.5W at idle) and they've happily run for years.


In other amps I've run them at 21W (cathode biased output stages) at 400V and they've also lasted for years.


The 6P3S is another story unfortunately. Completely different tube.

[HotBluePlates]

I've run the 6P3Se at ...

My experience is that those early straight-sided "6L6" tubes ...

I was in an Intro to Engineering class where we were divided into teams and had to design an 18" beam made of balsa wood which was required to support 250 pounds. The whole class was taught the relevant calculations and the different styles of beams. Part of the grading criteria included attaching a cost for materials (more wood = more $$$) and a cost for glue joints (more glue joints = more $$$).


Beams which passed load testing were then to be graded relative to each other based on lowest total cost. The thought was this would discourage an unsophisticated brute-force approach (make the beam 18" long but 3 feet thick).


On the day of testing, only 1 team had a beam that didn't twist and break under the 250lb load. My team's beam failed. We calculated everything correctly (so did every other team), but we placed too much emphasis on cost and gave ourselves almost no safety margin on strength. By comparison, the winning team's beam was about 3-4 times thicker than everyone else's (we thought they were silly and would get wiped out by cost penalties).


The point is the data sheets contain maximum ratings, but no manufacturer is going to make a 450v max part that break at 451v. They have to rate conservatively to allow some variation in operating conditions. They might even specify that you can exceed a maximum as long as you don't exceed multiple maximums at the same time (this is sometimes indicated under "Design Center Maximum" ratings).


Russian (and other foreign) tubes got a bad rap as fa as I know in the 90's based on reports of failures in Fender-style (specifically blackface) amps. This was when I first heard of "NOS"; at the time, I could still go into a TV repair shop in my city and buy as many used RCA 12AX7's as I could carry for a couple-bucks each. The Russian types we've been discussing were marked with U.S. type numbers (6L6, 6V6) and were failing in Fender amps that ran even the U.S. tubes way past their data sheet ratings. But the Russian types don't even have all the same high maximums as the American types.


So anecdotal evidence of getting away with this or that voltage is not something that's incredibly useful, in my opinion. I think if we go back to the source data sheets for the Russian tubes and run them according to their actual ratings (not by the ratings of the tubes we wish they were), that we could get good performance and as-long a life as any other tube.

[Jack_Hester]

Ok.  Here's the full sheet, plotted to an A0 size.  Makes for a higher resolution when zooming.

HBP -

If you would, take a look at the circuit that I have labeled with your mods, and see if I need to make corrections.  Or changes, as you see fit. 

Breadboard Update:

No work on the breadboard, this week.  Long days will dictate that. 

I did manage to rescue quite a few nice fuse blocks from the scrap heap, last week.  New controls I/O being installed for our new burners.  And, those guys gave me a heads-up on stuff that they were stripping from the Burner Control Building, that I might want.  I did (want, that is).  So, I've got small fuse blocks to put in series with my Line Feed to each transformer (PT, 3 x Filament).  Overkill, but what the heck.  I ordered some fuses from Mouser, though I managed to save a variety of sizes. 

No panels for projects, though the outage won't be over 'til the end of May.  I may be getting my hands on some scraps of shielded multi-conductor #16 cable.  It might be smaller, but I believe that it's #16-stranded.  Project parts. 

Anyway, I'll post progress as it happens.  Have a good one.

Jack

Update: See reply #23 for additions to the HBP PS mods.  No change in rev #.

[PRR]

> safety margin on strength. By comparison, the winning team's beam was about 3-4 times thicker
Clear spruce/pine/fir test-samples will take 3,000PSI. Code requires me to design as if it were good for 1,000PSI.
The classic spec on iron is 16,000PSI. Most iron will stand 50,000PSI.
We de-rate the material according to unexpected knots, cross-grain, splits, and slag which may be found in small areas scattered randomly in large pieces.
One reason wire suspension bridges stand-out is that wire has a higher strength (due to extra kneading) and "can't" have serious weak-spots (the wire would break in drawing).
You can also find more strength in wood if you split it rather than just push a saw through. This gets all grain going the length of the piece, and is prone to finding weak spots. Classic way to make small parts such as arrows and trenails, even up to ox-yokes.
My Dad once noted that *only* in electronics do we use 10%-50% safety margins. Houses typically run 15PSF on 3000PSI wood but are designed for 40PSF loads with 1000PSI stress in wood, two 3:1 margins or 9:1. Lifting rope is derated 3:1, and overhead lifts are often 10:1 safety margin.
WWJD? What Would Joseph Do? The guy Jesus lived with didn't have charts and calculators, yet everybody had a house or a stable or a manger, and obviously they didn't fall-down all the time.
Carpenters work by eye-proportion, adapted to whatever cubit or span or finger or inch units were in vogue or handy at the time.
The depth of a beam should be 1/10 to 1/20 of the span. 1/12 is a very safe number. 1/15 also works acceptably.
We might span a 12 foot room with a solid slab 1 foot thick. It works, it's costly, and on test it will support absurd weight.
To do better we need a saw. Sawing was expensive in Joseph's time, but later Romans had water-power sawmills. Transportation was still expensive. Today we saw and ship wood widely, so we can cut-up and cut-down wood for better proportions.
For distributed loads around 40 PSF, in common lumber, the width of each joist can be 1/10th of the spacing between joists. But then you have to fill between joists.
We find that our sawmill costs rise below 3/4" slices and we want two layers of floorboards so pennies don't fall through the cracks. 1-1/2" wood would span 18", or 20" OC joists, except half our floorboard run the wrong way for strength. 16" OC has worked well.
If joists are 16" OC then joists can be 1.6" wide. They were, until the late 1950s. Improved inspection became cheaper than the extra 0.1" of width so today we use 1.5" stuff.
If room is 12 foot wide we can pick 12'/15= 0.8' or 9.6" tall joists. Again due to the inspected-in strength of construction lumber we use "2x10" which is 9.2" tall.
Joseph didn't have inspection-graded lumber, but would have used about full 2x10s for important projects, 2x6 for cheap projects where the client couldn't complain about deflection or overload margin.
So... 18 inch span, 250lb load (distributed?). WWJD?
18"/15 is 1.2 inches, however the project is small and we might wish to go 1.5".
However the rules, while vaild for lumber and adaptable for steel, may not be good for Balsa. Balsa also has absurdly bad strength/cost ratio.
And here's my "WWJD?". Carpenter Joe would not use Balsa, even if KonTiki was abandoned on the beach. It's too bulky in the donkey-cart for the strength you get. Bulk is good up to a point (radius of gyration) but unless you are building a short squat Balsa Tree, and need internal water-storage (or KonTiki with short-term internal air space), it's really over the top.
Joe would sell the balsa and get a 2-foot 2x6 stud-scrap. Your bridges may have been $20 of Balsa? I'd sell you a chunk of 2x6 pine for a lot less than that. Without calculation, Joe and I both "know" we can stand on such a chunk, with tools, and feel quite safe; it would certainly stand a careful test at 250lbs.
However, given Balsa....
I find that the balsa-bridge load is often concentrated at the center. Ignoring local effects, most simple beams will handle 250lbs centered like 500lbs distributed. Over 18 inches that is 333 pounds per foot. If supported on residential-grade 40 PSF structure we need 8 feet of width, which is clearly unreasonable. Take out the 1:10 of joist spacing we have 0.83 feet width. Solid pine 10"x1.5" would work with 3:1 margin.
Balsa being 3X-10X weaker, this won't work. And that's quite a chunk of Balsa.
I had more but the forum is back to a 5,000 character limit, which really means 4600, and everybody else is already snoring....
Why is the post-box taking-out carriage returns for long posts in small font?

 

[HotBluePlates]

HBP -

If you would, take a look at the circuit that I have labeled with your mods, and see if I need to make corrections.  Or changes, as you see fit.

It all looks good to me.


Since you are breadboarding, and will try out multiple versions, try also re-including the 0.1uF bypass cap in my version across C7. Try without, with, and moved to across C9. I'd be very interested to find out if you note any changes.

> safety margin on strength. By comparison, the winning team's beam was about 3-4 times thicker
Clear spruce/pine/fir test-samples will take 3,000PSI. Code requires me to design as if it were good for 1,000PSI.

At first, the professor was horrified that she'd have to flunk 90% of the class for that 1/3 of the course.


Then she realized we'd applied the info she gave us, but there was no mention about what needed info we might lack, or that we should design with significant safety margins. Thankfully, she took it as a lesson for her as well as us.


In the 90's, I thought like everyone else that "Russian tubes are bad; NOS is good." No biggie, as the old stuff could still be found cheaply. That was the line spouted in print by largely non-tech writers who (I see now) couldn't really grasp what the issue was in imported tubes labeled by distributors with American type numbers.


Now that's not so much the case, so it seems like a good time to make use of the cheaper tubes.

[Jack_Hester]

Since you are breadboarding, and will try out multiple versions, try also re-including the 0.1uF bypass cap in my version across C7. Try without, with, and moved to across C9. I'd be very interested to find out if you note any changes.

Do you have something in mind, to listen for?  Tonal differences?  Noise filtering?  It should be easy to try just about anything that you have in mind, as these are very basic circuits.  Though I didn't include it, I tried a little mod that was offered for the first circuit.  It has, as built, has quite a bit of overdrive.  The mod adds another pot (Gain) and 22K resistor in series between the lug on the Volume pot, that the signal is tied, and signal ground. 
 
Jack

[HotBluePlates]

Do you have something in mind, to listen for?  Tonal differences?  Noise filtering? 

I'm not gonna put preconceptions in your mind about what you may hear, as that would be counter to the point of the breadboard.


Instead, I'd be interested for you to try it, and see if you hear any change. "No change" or "I don't hear anything" are valid results, and are just as informative as if you heard some dramatic change.


I haven't built a breadboard setup for myself as I don't have a dedicated workspace (like a garage or out-building) and have a 2-year old running around. I was intrigued by your board sub-assemblies, as I have a group of regulated power supplies that I could easily rig to such assemblies as a fast-setup/tear-down breadboard. So I'm using your experiment as a guinea pig until I get to try this stuff for myself. After all... you layout source brought up the notion of the film cap bypass of a filter cap, so why not taste-test it?


That, and I have an un-modded Valve Jr sitting right here because I found it cheap in a pawn shop. So I'm also gathering ideas before I actually open it up! 

[Jack_Hester]

I'm not gonna put preconceptions in your mind about what you may hear, as that would be counter to the point of the breadboard.

Fair enough.  There may be more people than myself involved, before the experiment is over.

Instead, I'd be interested for you to try it, and see if you hear any change. "No change" or "I don't hear anything" are valid results, and are just as informative as if you heard some dramatic change.

That will be easy.  But, a friend of mine from work can give them a proper test, and possibly make sound clips.  He has his own small studio.  I'll just haul it over to his house and let him have his way with it.

I haven't built a breadboard setup for myself as I don't have a dedicated workspace (like a garage or out-building) and have a 2-year old running around. I was intrigued by your board sub-assemblies, as I have a group of regulated power supplies that I could easily rig to such assemblies as a fast-setup/tear-down breadboard. So I'm using your experiment as a guinea pig until I get to try this stuff for myself. After all... you layout source brought up the notion of the film cap bypass of a filter cap, so why not taste-test it?

I'll make myself an outline of what needs to be tried, so that I can jot notes as they happen. 

I have an un-modded Valve Jr sitting right here because I found it cheap in a pawn shop. So I'm also gathering ideas before I actually open it up! 

It just so happens that I have what may be an early model EVJ, no cabinet.  I have never fired it up, so I have no idea if it works.  It has tubes, and was found on a shelf in an antique mall, with a bunch of old radios.  That would be worth trying, when the breadboard experiments are up and running. 

For now, these circuits will be a good start.  Quite possibly side by side on the breadboard, though I won't commit to such, at this time.  It's still coming together.  I'll post that progress over on my other thread.  Have a good one.

Jack

[Jack_Hester]

The only update to the drawing is to include the two trial uses of C10 (.1uf) in the PS.  One with it across C7, and the other with it across C9 (as in the original).  Wasn't necessary to draw, but a good reminder for me, when it comes time to put them to the test.

Jack

[Jack_Hester]

Ok.  I'm at it again.  In the short amount of time that I had this morning, I decided to do a no-brainer and sort some tubes.  Among those, I found some more packages of Russian tubes, and sorted those first.

I found quite a few more 6P14P's.  And, among the packages, I found some 5C3S/5U4G's (nice big bottles), and some 6N9S/6SL7's.

I looked up the spec sheets on Frank's site, and they seem to be comparable to domestic tubes.  However, I'd prefer to know from people who have first-hand experience with them.  Web reviews give a pretty high rating to the 5C3S tubes. 

Please comment on these.  Thanks.

Jack

[John]

I  have the Russian 6SL7s, and like them just fine. I've used those, the Russian equ. of 6V6, and Russian equ. of 5881. To my ears, they all sound fine.


FWIW. 

[Jack_Hester]

I  have the Russian 6SL7s, and like them just fine. I've used those, the Russian equ. of 6V6, and Russian equ. of 5881. To my ears, they all sound fine.


FWIW. 

Do you have the Russian number for the 5881's?  I'll be on the lookout for those, as I've got many more tubes to go through, including Russian.  Thanks for the comment. 

Jack

[John]

I  have the Russian 6SL7s, and like them just fine. I've used those, the Russian equ. of 6V6, and Russian equ. of 5881. To my ears, they all sound fine.


FWIW. 

Do you have the Russian number for the 5881's?  I'll be on the lookout for those, as I've got many more tubes to go through, including Russian.  Thanks for the comment. 

Jack

Jack, I can't swear to it, but I think they're 6P9SE; 63PS-e  I will check tonight when I get home. Tubeswell has used them and likes them also. I bet he knows the number off the toppa his head.


I have a pair of the TungSol 5881 reissues I got from Doug, and really like them as well.


*edit* I googled. The "e" designation is the better tube, handles higher voltage, and (subjectively) sounds better.... from what I read, anyway.

[Jack_Hester]

No, they are a good substitute for the 6L6G. Not 6L6GB or 6L6GC.

Look at plate dissipation first: 20.5w, which is comparable to the metal 6L6 and glass 6L6G rating of 19w.

From the plate dissipation, and from HBP's statement above, 6P3S-E tubes are more like the lower powered 6L6G's. 

Jack

[Jack_Hester]

Today was a day of much rest, and to try to shake the jet lag of rotating back to day shift tomorrow.  So, what time I wasn't napping or reading, I was drawing. 

Made a couple layouts for your inspection and comments.  Please indicate improvements, as layouts don't come easy for me.  Point-to-point is more in my comfort zone. 

Have a good one.

Jack

Update: Added BOM with Mouser Part Numbers.

[Jack_Hester]

I updated the drawing to have a BOM.  Unfortunately, it did not plot the same size as my drawings, so you will have to zoom in.  It is, however, a high resolution print.

Jack

[kagliostro]

The price of the 6P3S-E tube


http://tubes-store.com/product_info.php?products_id=55


is far from the price of the 6P3S tube


http://tubes-store.com/product_info.php?products_id=23

---


http://www.tubes.ru/techinfo/HiFiAudio/6p3s.html

K

[sean_thornton]

Seeing as you are sticking to Russian tubes have you tried the 6p1p-ev? This is very similar to the 6V6 but with a 9-pin base. They sound gorgeous in either SE or PP and are quite happy with output trannys used for el84's.
When i use 6p14p tubes in my commercial builds i use the ones with the suffix "-ev" . This means extended use for military applications, and they are my favorite el84 type. The only one i prefer to these is an NOS Brimar that i keep in my own very heavily modded EVJ (can it still be called an EVJ if the only thing i use are the trannys?).
I can post clips if you want them but i think most people on here have seen them.
Have fun with the Russian tubes, they can sound sweet if done right.

[Jack_Hester]

Seeing as you are sticking to Russian tubes have you tried the 6p1p-ev? This is very similar to the 6V6 but with a 9-pin base. They sound gorgeous in either SE or PP and are quite happy with output trannys used for el84's.
When i use 6p14p tubes in my commercial builds i use the ones with the suffix "-ev" . This means extended use for military applications, and they are my favorite el84 type. The only one i prefer to these is an NOS Brimar that i keep in my own very heavily modded EVJ (can it still be called an EVJ if the only thing i use are the trannys?).
I can post clips if you want them but i think most people on here have seen them.
Have fun with the Russian tubes, they can sound sweet if done right.

I have a grab-bag (so to speak) of quite a few more Russian tubes that will take a while to go through.  But, I believe that I saw some 6P1P in there, though I don't remember if they had the EV suffix.  Now, I'm going to have to check when I get home.  Thanks for the heads-up on another to be on the watch for.

Jack

[Jack_Hester]

The price of the 6P3S-E tube

http://tubes-store.com/product_info.php?products_id=55

is far from the price of the 6P3S tube

http://tubes-store.com/product_info.php?products_id=23

---

http://www.tubes.ru/techinfo/HiFiAudio/6p3s.html

K

I got my 6P3S-E cheaper than his 6P3S's.  I've had them around for awhile.  Got a hold of most of my Russian tubes, a few years ago, when I decided to make my return to tubes.  Didn't know what I was buying at the time, but cheap was the key word.  Figured I'd use them for something.  Now, I can. 

How do you get to the tubes.ru site to look at other tubes?  Every way that I've tried, so far, has told me 'Forbidden'.  Thanks for the listings. 

Jack

[kagliostro]

Try this link
http://tubes-store.com


if you are not able to connect, try to use the openDNS, setting it on your PC

208.67.222.222
208.67.220.220

instructions to be used if you don't know how to do
https://store.opendns.com/setup/computer/

The same instructions can be used to set the Google DNS that you can try in alternalive

8.8.8.8
8.8.4.4

if that didn't resolve the problem, tell me which tube to search there and I'll search it for you

Ciao

Franco

Here you can see some Fender layout converted to 6N2P tube
http://www.tdpri.com/forum/shock-brothers-diy-amps/335898-5c1-5c3-5f1-5e3-concurrent-builds-6.html

[Jack_Hester]

Thanks for the reply, Franco. 

I'm trying to get to the last link that you have posted, to look at more of their tube data sheets. 

Jack

[HotBluePlates]

Or go here. Or go to Frank's and click the Search button in the top right. If Frank's doesn't have the data sheet, you probably don't need it... 

[kagliostro]

Jack I apologize

I misunderstand you, I was thinking you were looking for other russian tubes, not for their datasheet

to find those datasheet I simply use Duncan's search engine

http://tdsl.duncanamps.com/tubesearch.php

Sorry for the inconvenient

Franco

[Jack_Hester]

Or go here. Or go to Frank's and click the Search button in the top right. If Frank's doesn't have the data sheet, you probably don't need it... 

Thanks.  Frank's has been one of my primary tube data sheet sources.  I was curious as to the source that Franco gave.  I now have it bookmarked.  I usually check more than one source, just in case.  You're right about Frank's.  It's hard not to find what you want on his site.

Jack

[Jack_Hester]

Jack I apologize

I misunderstand you, I was thinking you were looking for other russian tubes, not for their datasheet

to find those datasheet I simply use Duncan's search engine

http://tdsl.duncanamps.com/tubesearch.php

Sorry for the inconvenient

Franco

No apology necessary.  Sometimes, I don't ask the question correctly. 

Got your link bookmarked.  Thanks for the source.

Jack