KT88 SE Abdellah (modified by Alex Gendrano) + AK..

[Kegger]

Mini Mod..

Just doing last minute updates to a new build an thought I'd post the schem.
(same basic amp as the thread)

I wanted to make the amp more receptacle to using the 6N1P an or the 6DJ8.
(without swapping LED's around with a switch and better voltage for 6DJ8)

(to set it up initially)
Basically was 1 2v red LED then a 100ohm resistor below it, a bit of feedback
this way (just a touch), and lowered that plate load a bit just to "tweak" the
circuit to work well with both the 6DJ8 and the 6N1P, also the 6BK7 family of
tubes, plus if more gain is needed/wanted the ECC85 can drop in here as well.


The thing I'm trying to accomplish with the LED is minimal feedback so we get
as much gain from our driver as we can an not use a cathode bypass cap here.

This is shown as an amp with no volume control, using a separate preamp, the
amp can be done just like the original if you want volume control, or converting
the original to no volume control wire it up like this input stage.

Also..
(updated cathode resistor on output tube for better current, both schematics)
You could run the driver with just a 2.8v led, but the unbypassed resistor helps.

Cheers..

 

[zekk]

6CG7 at frontline!

Rolled in the 6CG7 in place of 6DJ8.
It works and looks better (taller) , but sound wise, ...... :scratch2:.
Needs to listen further for a better judgement.

What puzzled me is that it was not included in the Tube Handbook as a direct sub for 6DJ8!

WOW! What a journey!

 

[Kegger]

Yah the 6CG7 will "work" in place of the 6DJ8, but it has much higher plate resistance
an less gain then the 6DJ8, so it won't draw as much current or drive the output tube
enough in this circuit to get full power. (hence why it's not considered a 6DJ8 sub)

 

[AmadeusMozart]

Just doing last minute updates to a new build an thought I'd post the schem.

.....

Kegger

I received all the iron from James in Taiwan. With it came documentation as well.

Don't know if you are aware of it but the OPT is rated as 90 mA @ 2.5 k, 80 mA at 3.5 K and 70 mA at 5 K.

From the cathode resistor and the voltage at the cathode I conclude that you are running at 78 mA and it may well be that this is at the upper limits of what the winding itself can cope with. (Would hate to see you burning out that transformer.)

Am working on the chassis - shipping something was just out of the price range and I have a aluminium wholesaler within a few Km so have been going there to get pieces and have been gluing those together with JB Weld. Will add some countersunk screws to ensure electrical contact (JB Weld is not conductive). If other want to folow this avenue: make sure that the aluminium is totally grease free otherwise it will not stick over time. I do this by soaking the aluminium in boiling hot water with lots washing soda for about 10 minutes (basically a form of anodizing) and then rinsing it under the cold tap and drying this with a clean tea towel.

Once finished I intend to put up some pics of the different stages and the finished product.

AM

 

[Kegger]

Sorry, hadn't updated voltage on outputs cathode, also along with it's resistor..
(current should be a bit lower now according to the schems)

But at the same time I did not know the limitations on the James outputs, thanks..
(I tend to run Edcor more then anything else, I got the spec from them at 100ma)

------------

(Also this was posted earlier)

"FYI, I built the Abdella SEP amp last year and discovered he has a small error in the original schematic.
(and continues onto the Gendrano version here) I used the suggested James 6123 OPT and discovered
when set to Z=5K the UL screen tap is actually at 35% not 43% (It would be 43% when using 3.5K taps)
HTH WopOnTour"

--------------------

So I would suggest when running the James use 3.5k taps, more then 1 reason now..

 

[AmadeusMozart]

What about using a 6L6GC-STR from Tungsol?

Kegger

I've just finished the RH807 and discovered something about the g2 voltages. Basically the g2 can go in reverse current and when using a 10k resistor to drop the voltage some things start exceeding specs.

Then in the KenRad specs (IIRC) it is mentioned not to use a resistor to drop the voltage to g2. Tubes in the past were constructed under different tolerances than today and with a beam power tube the location of the g2 wires is critical for its performance. (and very sensitive to any deviation from its intended location)

I wonder if I can run your design (UL) with a 6L6GC @ 350V and still get respectable performance (don't want to start experimenting, rather use a proven design). Since I have some 500 ohm / 12.5 watt rheostats with a 120 ohm resistor in the cathode I should easily be able to adjust the current through the tube.

(what happens when we drop the voltage and drop the impedance of the OPT at the same time, e.g. going to 2.5 K?)

Thanks, AM.

 

[Kegger]

I'm not exactly sure what you have built and what your asking.

What voltage is your power supply, what output tubes are you running now?

I would have no problem running a cathode biased 807 or 6L6GC amp at say
a 425v supply or less, (maybe 450v) probably wouldn't go below 350v though.

 

[AmadeusMozart]

I'm not exactly sure what you have built and what your asking.

What voltage is your power supply, what output tubes are you running now?

I would have no problem running a cathode biased 807 or 6L6GC amp at say
a 425v supply or less, (maybe 450v) probably wouldn't go below 350v though.

I've built Alex Kitic's RH 807 (see attachments) with a B+ of 350V and am using James JS-6123HS OPT set to 5k. Your KT88/EL34 SE design is using a 3.5K Zpri and 450V. Since I was not happy with the fluctuation g2 voltage(going negative at times and resulting in a fluctuating cathode voltage indicating that the tube was not working within manufacturers specs / recommendations) I added a flyback diode and a zener from B+ to the screen grid (g2) with a leakage resistor to ground. This solved the problem.

However I do not consider this to be the only solution and was looking at your KT88/EL34 amp as an alternative but the voltage is a loot higher than I can achieve with tube rectification with a low enough ripple.

My questions are: Can I drop your design to 350V and if using the 6l6GC-STR from Tungsol (plate dissipation 30W and screen dissipation 5W max) can I drop the OPT primary impedance to 2.5K and still get around 8 watt output?

Thanks

 

[Scuzzer]

I added a flyback diode and a zener from B+ to the screen grid (g2) with a leakage resistor to ground. This solved the problem.

Just curious, the 807 spec sheet says not to use a dropping resistor but instead use a voltage divider. Did you try this as well?

 

[Kegger]

AM, with a 350v supply you'll get in the neighborhood of 6w, no problem running
the circuit with your voltage and or your output tranny, and a 2.5k to 5k output
tranny will work in this circuit here, so yah give it whirl! (807 and or 6L6GC)

 

[AmadeusMozart]

Just curious, the 807 spec sheet says not to use a dropping resistor but instead use a voltage divider. Did you try this as well?

A voltage divider needs more idle current to keep the fluctuating current in g2 under control. I have ordered a OB2W gas stabilizer (108V drop) which will replace the zener. There is less noise on tube voltage regulator than a zener and the heat stays outside of the inside of the chassis making things run cooler inside. Plus it looks nice too!

The dropping resistor will be there for ensuring that the voltage does not rise but I may consider a CCS in there as well.

Once set up I will check out the sound difference between the UL and the pure pentode operation and implement permanently what I like best.

AM

 

[Scuzzer]

Once set up I will check out the sound difference between the UL and the pure pentode operation and implement permanently what I like best.

AM

I breadboarded the RH807 and then changed over to something similar to the simple SE to check out the differences. On my substandard breadboard output iron I liked UL better but that might change if I ever get better iron. The 807 screen rating kind of limits the UL power (as I'm sure you know) so I'd be interested to see your final schematic if you decide that pentode works well. Or are you thinking of switching to 6L6s and keeping UL?

 

[AmadeusMozart]

I breadboarded the RH807 and then changed over to something similar to the simple SE to check out the differences. On my substandard breadboard output iron I liked UL better but that might change if I ever get better iron. The 807 screen rating kind of limits the UL power (as I'm sure you know) so I'd be interested to see your final schematic if you decide that pentode works well. Or are you thinking of switching to 6L6s and keeping UL?

On the "other" forum there is a thread about the Schade feedback. In a nutshell (as far as I understand): triode's sound better (unassisted by feedback) than pentodes due to the high internal impedance of pentodes and the low internal impedance of triodes. Schade feedback can bring the pentode internal impedance down to a triode level or even lower (depending on how much feedback you use). As a result a pentode can sound just as good as a triode. (or even better)

UL came around after the 6L6 / 807 was designed. As far as I know the 6L6 was designed for push pull operation with lots of h2 (which gets cancelled out when running in PP) and very little h3. Having said this: in SE the amount of h2 and h3 depend a lot on how much you are trying to get out of the tube. The tube was designed with an optimum g2 voltage of 250 - 300 V. Another item not to forget that the 807 / 6L6 are beam tetrode's and they react differently to UL than a pentode like the EL34. One can build a tube amplifier with so much feedback that it starts sounding like a SS amplifier and I do not want to have a SS sound.

At this moment my system looks exactly like the RH807 except that instead of using the 10k resistor from B+ to g2 there are some zeners with a film bypass cap of 10nF and a 50K resistor to ground. The junction of the zener/10nF and 50k resistor is where the g2 is connected through a 100 ohm resistor.

I'll be replacing the zeners with an OB2W (108V gas tube regulator) since it a) looks nicer and b) dissipates the heat above the chassis and c) keeps solid state out of the amp and e) is more temprature stable than a zener.

One other note: the sound of the 6L6G(C-STR) changes dramatically around 18 - 20 W anode dissipation. (ask the guitar bro's :yes: ) so you might want to do what I did: instead of having a fixed resistor in the 6L6GC cathode put in a 12.5 watt rheostat of 500Ohm with a 2W 120 Ohm resistor in series. You can then adjust the current easily.

AM

 

[Scuzzer]

On the "other" forum there is a thread about the Schade feedback.

Yes, I have read that thread. I understand the purpose of that type of feedback but I just haven't gotten around to buying the parts to fully investigate the possibilities. Thanks for the information. A nice glowing VR tube is always preferable to SS stuffed under the chassis especially if it accomplishes the task better.

 

[AmadeusMozart]

Yes, I have read that thread. I understand the purpose of that type of feedback but I just haven't gotten around to buying the parts to fully investigate the possibilities. Thanks for the information. A nice glowing VR tube is always preferable to SS stuffed under the chassis especially if it accomplishes the task better.

PS I found on eBay these rheostats at a very acceptable price. Liked them so much that I got even a second pair, just in case.

AM

 

[Kegger]

Also remember yes for pentode operation that screen is rated for 300v but in triode
it is 400v, so with UL and the 1.2k screen resistor from the UL tap I treat UL with a
triode rated screen voltage, so 400v should/would be safe, so take say 450v supply
with cathode biased 807, loose say 10v through the output tranny, then say 40v on
the cathode for bias, now your down to 400v plate to cathode, an 807 runs that all
day no problem, I've ran um there an they like it just fine.

-------

As far as feedback goes, generally speaking a triode won't need much if any feedback
to have low output impedance, linear output and a nice distortion spectrum, Pentode
really does need a decent dose of feedback to reach those same capabilities, take an
pentode an run it UL with just dash of feedback will go a long way to excellent results.
That is the circuit here that I've been running for some time now on various pentodes
with just a dash feedback (Schade type) from the plate of the output back to plate of
the diver so ends up plate to grid on the output tube, not much feedback though with
a 270k resistor being used in this setup. (try 100k - 270k for the feedback for taste)

This is the circuit an amps I've been experimenting with past few years now, I like what
I get from this on large variety of tubes, it just works very well in the amps I've built.

------------

Not saying that others shouldn't experiment, just pointing out this is a seasoned circuit
that has had a lot of testing on it's basic design and has come though very very well.

 

[Scuzzer]

Yep right around 400V is where I was planning to run it in UL. My current breadboard is only good for about 350V Plate to Cath so I don't want to make any firm decisions until I get some iron for both ends (power and output). Thanks for confirming I'm not going too far off the reservation.

 

[Kegger]

I'll be curious with both of you guy's experiments, please post along as you go..

 

[Scuzzer]

Mines not going to be anything fancy. I've already got all the parts on hand to put together a Simple SE 807 version minus the iron. I figured I'd start there and then play with it. Here's my goofy oscilloscope chassis "breadboard" with a 5933 in it.

 

[Kegger]

I love it, looks almost as good as my latest prototype amp.:thmbsp:

.
.