diytube.com

Here are a couple of my more recent Clementine builds:

Full frontal tubity:



and under the skirt:



PSU filter string is CLCRCRC.

Built into and onto a 10X6X2 chassis, it was a "sardine can" fit
OPTs are Transcendar 5K:8 ohms, 5W, bought from the fee-bay place.

Now here is Clementine's Chubby, 6L6GC (actually, Russian 6P3S-E / 6n3C-E) triode-strapped brother, "Ivan the Terrible":

Full frontal tubity:



Under the skirt:



PSU filter string is CLCLCRC

Overhead "spy satellite" view:




I had to get a bit creative in order to use the 180 uF/500V caps, and a shallow "dumpster" chassis depth of 1.5 inches (38.1 mm).

I also built a near-twin to this Chub, but set up for 12V6GTs, DC filament operation, B+ is 325VDC. The 12V6GTs are in triode-strapped operation.

I'll add its pics RSN.

They sound great!

I have two more Clemmies in progress, the first will use 12A6s, the next one, I'll mod for 25E5s (with an 82 ohm 5W resistor in series with the 12SL7 filament).

Hearing great sounding music can really have a rather profound effect on the listener





/ed B

I had to get a bit creative in order to use the 180 uF/500V caps, and a shallow "dumpster" chassis depth of 1.5 inches (38.1 mm).

You must live in an upscale neighborhood. None of the dumpsters arround here ever have anything to use for a chassis. Although I once found a baby stroller I though might have some potential...

20

The dumpster was located at my last job, these were "rejects" because they had minor manufacturing errors, and couldnot be used, they simply ended up in the scrap metal bin, and there were literally hundreds in there, in many different sizes and layouts, most of them, unfortunately, not readily usable without major "surgery" because of internal welded shield partitions.

Many were seriously heavy gauge steel, and a real beast to do any drilling and punching, but they probably could have survived being run over by a bulldozer. THough, I say, never say never - I took a few, just in case I decided to attempt make a "zombie apocalypse" amplifier

/ed B

EWBrown wrote:The dumpster was located at my last job, these were "rejects" because they had minor manufacturing errors, and couldnot be used, they simply ended up in the scrap metal bin, and there were literally hundreds in there, in many different sizes and layouts, most of them, unfortunately, not readily usable without major "surgery" because of internal welded shield partitions.

Many were seriously heavy gauge steel, and a real beast to do any drilling and punching, but they probably could have survived being run over by a bulldozer. THough, I say, never say never - I took a few, just in case I decided to attempt make a "zombie apocalypse" amplifier

/ed B

I've saved some desktop computer cases from back in the day they were built shield out Gamma rays and hold 50 lb monitors. Now I just need to get the steel cutter blade to use with a circular saw to rip them down to size.

Those are all still in great condition, nicely enammeled. I have my eye on some old Microwave cabinets to try down at the GW, too.

As promised earlier, the 12V6GT triode-strapped "Pleasingly Plump" Clementine

Basically the 6L6GC "Chubby", but with 12VDC filaments and approx 325 VDC B+ at the OPT primary.

Power trannie, older Musical Power Supplies 285-0-285, 120 mA, plus 2 6.3VAC filament windings, good bargain-priced power iron at the time. OPTs are GXSE10-8-5K.







/ed B

Nice. How'd that overlapping hole punching go for the PS caps?

With the aluminum (it is a different alloy than the hammond chassis) the overlapped punching was relatively easy, I've tried it with steel, and made a mess of it. After a few previous experiments,with scrap metal, it works best to punch them in a 1- 3 - 4 - 2 order, starting from either end.

These "dumpster chassis" are thicker, and perhaps a bit "softer" alloy than the usual alum chassis, the metal punches and drills very easily and cleanly. With the extra thickness, they are very strong.
Welded corners, not just folded over flaps with tiny spotwelds.

These were originally intended to house a fiber optic network converter, in an RF-tight box, for certain "special purposes" that certain unspecified "nonexistent agencies" might have been using at some time in the recent past

I also decided that a 180 uf cap wasn't needed at the VA end of the PSU CLCRCRC filter string, 47 uF was more than sufficient.

/ed B

Just finished the third Clementine.. Originally started out as an experiment, tried using 12A6s in UL SE mode, with "1626" B+ voltages, and with Triode TF103SE-48 OPTs (7500 ohm primary Z, 350 0hms DCR). and 500 ohms (2X 1K, 2W in parallel) cathode resistors,and the NFB loop, similar to that used with the 6L6GC circuitry.

It sounded really, really, really bad... thin, sharp and fizzy, and with very low power output, probably about 1/4 watt at best.

OK, 12A6s are out... They're best left to do push pull duty, rather than SE. I'm guessing the internal plate resistance is just way too high, for SE, even though the tube data states they are good for 3.8W @ 10% THD in SEP mode, with 250V and 7.5K OPT. In your dreams....

Next, I tried 12V6GTs as they have the same pinout, and are pin-compatible.
Somewhat better sounding, but still fairly low power output, 12V6GTs definitely like having a higher B+ voltage and cathode current. 325V and 45 mA is more to their liking than a relatively "feeble" 250V and 25 mA.

OK, clip out one of the paralleled 1K cathode resistors for each V2 and V3, and take out the NFB loop, and plug in a couple of 1626s.

Ahhhh, much better... Even with the small cheap Triode SE OPTs it sounded pretty decent, not quite as "full and rich" as with the better (and more costly) Transcendar OPTs, but still far better than with 12A6s or 12V6GTs with 250V and run at approx 27 mA cathode current.
The 7500 ohm primary lowers the maximum power output, probably closer to 500-600 mW rather than 3/4 watt.

For the fouth Clementine board, I'll try setting it up as a floating paraphase PP single channel, with 6L6GCs and 6SL7. Should be a pretty simple mod, with no etch-cutting needed, just re-routing a couple of resistors, and adding a couple jumper wires. Stay tuned...

/ed B

EWBrown wrote:
For the fouth Clementine board, I'll try setting it up as a floating paraphase PP single channel, with 6L6GCs and 6SL7. Should be a pretty simple mod, with no etch-cutting needed, just re-routing a couple of resistors, and adding a couple jumper wires. Stay tuned...

/ed B

Hi Ed ,

I have three clemintine boards here and was going to do someting similar with two of them only using a bridging input transformers .Haven't decided though to make a power amp or a preamp . For the power Amp would use a pair of UTC LS-12 input transformers and UTC LS-57"s .For a line stage Preamp , was planing on using a pair of UTC a-18' and A-26's as outputs , and use 6J5GT's for the output tube which have the same pin configuration as 1626's .

A 10k to 15k input transformer like Edcor makes would also work for an input transformer Would be a much cheaper alternitive to the UTC's


http://www.edcorusa.com/p/95/ttpc15k-10k Use 10K in To 15K out
http://www.edcorusa.com/p/164/xsm15k-600
http://www.edcorusa.com/p/367/xpp1-600-15k

EWBrown wrote:Just finished the third Clementine.. Originally started out as an experiment, tried using 12A6s in UL SE mode, with "1626" B+ voltages, and with Triode TF103SE-48 OPTs (7500 ohm primary Z, 350 0hms DCR). and 500 ohms

/ed B

Ed,

i love following your tubal libations...

Did a search for that tranny and came up with 5K Z on the specs. Is the TF103SE different from a TF103-

I have the floating paraphase (also known as self-balancing phase inverter) board built up, setup for 12V filaments (12SK7 and two 12V6GTs). Haven't yet applied powe rto it, I'll try it out, in another day or two...

I'll describe the relatively simple mods, once I ascertain that they make fora decent sounding amplifier. If it does not, then it's a very simple and quick matter to revert it back to 2 channel SE/UL outputs.

This should also work well with the G*S*G boards.

Circuit is similar to this, except that the 500K resistors become 220K, 1% resistors, and the 6SN7 is a 6SL7, with 100K plate resistors and 1K cathode resistors.



As the ARRL handbook describes the circuit's operation, and I've grossly simplified it here, it works by maintaining the junction of the three (500K) grid resistors at zero volts, as the output of one side goes positive, it forces the lower driver triode's output to an equa,l but opposing voltage (in theory)....



/ed B

BTW, that tranny must have multiple incarnations. I found the docs on it as a 5K at one time and now they have it listed as an 8K...

I had run a "ratio test" on the 103s that I used and they worked out pretty close to 7500 ohms, and, yes, the earlier ones were 5K.

/ed B

EWBrown wrote:I havethe floating paraphase board built up, setup for 12V filaments (12SK7 and two 12V6GTs). Haven't yet applied powerto it, I'll try it out,in another day or two...

I'll describe the relatively simple mods, once I ascertain that they make fora
decent sounding amplifier. If it does not, then it's a very simple and quick matter to revert it back to 2 channel SE/UL outputs.

This should also work well with the G*S*G boards.

Circuit is similar to this, except that the 500K resistors become 220K, 1% resistors, and the 6SN7 is a 6SL7, with 100K plate resistors and 1K cathode resistors.



/ed B

Ed,

I've seen this circuit and a similar one that grounds the inverted output near the common 500K ground point, before it goes back around into grid again.

So in this circuit what keeps the signal from looping back and creating a real mishmash signal rejoining the output from the first tube?

The Valve Wizard explains it better than I...

http://www.freewebs.com/valvewizard/flo ... phase.html