• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

The "$250" Amp? - 6LU8 Compactron SE, UL

I've noticed that using a better driver than the 12AT7 actually resulted in more distortion, so some cancellation must be going on there that give the amp it's 0.8% THD at 1 watt in triode. With a 6H30 w/ CCS load (very linear), the amp rose to 2.5% THD at 1 watt - so the 12AT7 is quite suited to this. Hopefully the triode in the 6LU8 will do the same for you guys, the curves look somewhat the same, but a tad more linear than the 12AT7.
 
I've got all parts needed in hand, with the exception of the PT. It took a dozen calls to Antek, but they assure me that they've finally shipped it. I'll believe it when I see it, but what can I do? The lack of PT and the SMAC meet will push the initial build off to next week, unfortunately.

Bricktop, that's an interesting discovery to make about the 12AT7 and distortion. The original designer of the RH series of amps said that he arrived at the 12AT7 after a lot of experimentation, that it was no accident. Eli Duttman uses the 12AT7 as a driver in his "El Cheapo" amp, as well. I think that both designers say that it has a very complimentary distortion. I myself found that driving the RH schematic with a 6AU6 was a bad mistake, and when I switched to the 12AT7 everything sounded much better.
 
In the schematic C2 is an electrolytic, bypassed with a better quality cap.

If one had a nice poly they wanted to use there instead would be an upgrade.

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

Hey Bricktop if u still have your prototype together could u try the praposed
setup with a 300v supply, UL an the 100K P-G feedback scheem in place to?
(Also I'd be curious as to what the -bias voltage is to achieve 40ma)

I'd be interested in how that setup works out on the 6AV5..
 
Kegger said:
Hey Bricktop if you still have your prototype together could try the praposed
setup with a 300v supply, UL an the 100K P-G feedback scheem in place to?

I'd be curious how that setup works out on the 6AV5..
Click to expand...

Lemme think about that... I put the transformers in an amp, I have to think how hard it is to sub in the 6AV5 where I got 6JM6s in there now... I'll likely be setting the amp on a side with tubes hanging out :D I'll check when I get home tonight. 6JM6 is a stretch from the 6AV5 or the 6LU8 pentode, I just had a bunch of them... Changing the power supply voltage in it won't be an issue, it has a fully regulated supply built in with a 6080 pass tube and 12AV7 error amp. How much current did you want to try?
 
Step ahead of yu, I modded my post just before yu posted, ;) 40ma per pentode. (to start, will increase later)

Thanks, if it's a big deal don't worry about it,
Keg
 
Kegger said:
Step ahead of yu, I modded my post just before yu posted, ;) 40ma per pentode. (to start, will increase later)

Thanks, if it's a big deal don't worry about it,
Keg
Click to expand...

At 40mA, ya won't be able to get more than 2 watts out of it - it'll be going into cutoff with a 2.5K load (is that what you are using?), I'll push it a little harder. At about 65mA, near 5 watts should be doable. It's about 20 watts plate, but should be OK from what I've heard from tubelab. 20 watts may be OK for the 6LU8 pentode, after I test for bias voltage, give it a bench test and peek for glowing screen wires or plates. I think the best best is to plant the idle where the maximum voltage swing and cutoff current have similar power limiting effects, probably somewhere around 250-300 volts and 60-80mA. If the output power is limited by bias current, then there is no advantage to running a higher voltage (aside from perhaps some distortion specs, but I think we are going for max power from this design right?), and vise versa.
 
(I here what yur saying Brick, we probably will end up closer to 50ma here)

Well the problem I have is there are no specs for a 6LU8 or 6AV5 in UL, So I
want to see what it will do at 300v and 40ma, we have both 2.5k and 5k UL
outputs to use on this setup. And were not necessarily going for max power
but we will be experimenting with different loads and operating points to see
what we can get from the setup, also I'm trying to get near 300v so I get a
good operating point for the driver and still be able to use the P-G feedback.
(And yes triode strapped for the output will be used as well)

I also believe the 6LU8 is quite the different animal then the 6AV5 is here to.

I just don't want to jump to conclusions from looking at curves that have no
real world testing done to determine the operating point. Also in our setup at
300v it will be cathode biased so the plate to cathode voltage on the pentode
will be say 275v or so but the driver will still get the 300v supply plus the fbk it
will add pretty much another 25v available for the driver to use.

A first inclination is try 10.5 to 12.5w plate dissipation, so 40-45ma would be at
about what I'd be shooting for to start the testing, and increase from there out.

I'm not planning on this thing delivering a good amount of power, and will be very
conservative to start, then see how far we can push them and what we get out.
Most TV sweep tubes can run quite a bit passed what they can do in TV setups
when there used in Audio were there not beat up as much, so I'd suspect yes we
can get quite a bit more from them compaired to where were starting them out at.

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

(If we do want to ponder info)
By looking at the curves, voltages, currents, plate resistance and transconductance
this thing is closer to a 6l6g or el34 then it is the 6av5, the 6av5 will draw a lot more
current and need more biasing then the 6LU8 will.

Most of the sweep tube curves an measurements are done with TV setups in mind so
the pdf's only tell part of the story as to how the tube will run in Audio circuits.

So I'm not really sure how much help 6AV5 characteristics are going to be on this one
but the testing info is always good in my oppinion and we may learn something.

When we get the 6LU8 fired up and take some measurements we can compare how the
2 tubes react and then the 6AV5 info will be interesting to see how far off it truly is.

I can't find my notes now, but I did some voltage testing on 6LU8's wired in triode a while
back an came up with some biasing info to help build a tube model for spice to at least be
able to setup a cathode bias resistor value to get correct currents and fixed bias voltages
to get the same values I had tested, there quite different from the 6AV5 when I built it.

And consequently the drive needed for the 6LU8 will be much less then what a 6AV5 needs.
The 6AV5 is quite a bit like the 2A3 "when triode strapped" and does need a fair bit of drive.
 
I wasn't really considering any tube data when talking about the 2 watts out with 40mA and a 2.5K load. Tube choice and operating mode will only determine how it biases and interacts with the loadline. But, to get X watts out a single-ended load of R ohms, one needs at a very minimum 1.414*sqrt(X/R) amps of bias to keep the peak current swings from making the output tube go into cutoff, because there will be I^2*R AC RMS amps though the load, and sqrt(2) times that peak AC amps.

For a 2.5K load, 40mA will get about 2 watts before cutoff. For a 5K load, it will get about 4 watts - and these numbers are only relevant if the output tube isn't running near 0Vg1 territory - but at 300V that shouldn't be an issue...yet.

For 300V supply, the 5k load should be used (I don't have one to test), and in my head I think it should be biased at around 45-50mA to keep it away from cutoff. I'm guessing that the output tube's plate can swing down to, say 75 volts before vg1=0V (Might be able to go lower in UL, in which case the bias current will go back to being the limiting factor in power - so that may want to be increased again if the tube can take it). Approx 5 watts output should be doable with 300V*45mA = 13.5 watts plate dissipation - should be no sweat for that tube. I think you'll wanna be able to get all the power you can squeeze out of it for people that don't have super efficient speakers.
 
All good info Bricktop, and my post wasn't really aimed at you per-say, but more
for others in the thread who might be trying to decipher the data, at least some. ;)

And I just wanted to point out the 6AV5 and the 2.5k output may yield quite different
results then what we get with the 6LU8 and 5K will end up yielding us in the long run.

Yes with a 2.5k output and the 45ma we could not get the output power we can in
the 5k setup form, but if someone did have the outputs and wanted to build an amp
they could build one, just at a lower full power output/rearrange there voltages.

Looks like we are close to the same page with 5k outputs, 300v supply and 45-50ma.
Agree the more power we can get let's get it, just wanted to clarify it wasn't an main
goal of the project. At this moment with cathode bias the plate to cathode voltage is
roughly going to be 275v, so 275v x 45ma gives us 12.375w, yes with room to spare
I think as well, we should be able to stretch that out.

I was/am curious how the 6AV5 would perform in UL with the 300v supply and the P-G
feedback, if anything else just to see lower wattage performance of how the circuit did.
Yu know just a pre teaser of the complete curcuit in action so to speak. :)

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

Early on in the thread the specification was to use the 5k Edcor output, but with ED
and Bricktop having the 2.5K's on hand, so the testing was for more info, ED is gunna
build the protoype with the 2.5k then when it comes to me I'll put in the 5k units I have.

So yes still at this moment the 5K output is the intended output for the project and what
"should" yeild us what we were going after.
 
I might be sending something back to edcor real soon, so when they send me back the right product, I may throw a pair of XSE15-8-5K in there. Nice to have experimenting stuff around the workbench. If it works out that I get them before the proto is made and all that, I'll let you know of the results. :thmbsp:
 
Glad I just perused the last page of this thread. I'd hate to waste half an hour on 13 pages and still have no freakin clue what language you all are using. :D
 
Actually if you do decide to read the entire thread, ED and others have done a
fine job of keeping the techno babble to a minimum or at least explaining it as we
go along so most anyone can get a grasp if they are at all the least inclined to do so.:thmbsp:
 
bricktop said:
I might be sending something back to edcor real soon, so when they send me back the right product, I may throw a pair of XSE15-8-5K in there. Nice to have experimenting stuff around the workbench. If it works out that I get them before the proto is made and all that, I'll let you know of the results. :thmbsp:
Click to expand...

Sounds great Brick, when ED and I were first talking this project (and other projects) we
got the 5k 16ohm and the 5k 8ohm outputs so we had a multitude of options for "future"
things we may want to build/test, I'm gunna get the 4 ohm version as well to cover it all.

With having all 3 of those and different speaker loads that's a pretty decent range of cover.

As inexpensive as these outputs are and there track record, it's quite nice to have them.:thmbsp:
 
I never used EDcor before-but after reading the DIY link-great ex/otputs-one could use an upgrade further down the road-Im very pleased so far.
 
Kegger said:
Actually if you do decide to read the entire thread, ED and others have done a
fine job of keeping the techno babble to a minimum or at least explaining it as we
go along so most anyone can get a grasp if they are at all the least inclined to do so.:thmbsp:
Click to expand...

My goal is also to present the information in a step-by-step informative way, deciphered into straight English, with the design decisions taken as well as the reasons why and what the results were.

None of this can happen until we have a prototype, then a BoM, a final schematic, and then a production build, but I will begin taking notes, photos, picking brains, and etc, as soon as we start the prototype build.

My dream, if you can call it that, is to be able to present a virtual kit - not just a list of parts and a schematic, but the whole enchilada for the absolute beginner - even down to what is a resister and why do I need one. I don't expect you guys to speak English, just put up with me asking over and over until I can translate it into English well enough.

Don't anyone be scared off by the bias voltages and photon guns set to stun or whatever. We'll put this gobbledygook into some kind of order before we embark on the mass build.
 
Wigwam, count Zenith2134 in on this for sure. I just checked my tube caddy and I have at least 6 or 7 6LU8s, GE brand. I gotta take em to my friend to test them as I don't have a tester here, but I am excited for this amp.


As for extras, we can just add/remove as we please once we finish our build which will be based on the agreed-upon schematic. :thmbsp: That way everyone's happy. I just got my partsexpress catalog today too, and that got me in the mood to order up some stuff. :yes:

I just finished reading through the thread and I can say you guys have all got some great ideas. If I may add my two cents, I have to stand by Edcor's reliablity and sound for the most part. I never DIYed with their transformers but the ones I've heard were great.
 
tubesandmore.com --look for triodestore.com-also for parts I have done bussness w/patrs express for years and have had a great ex with them.
 
You must log in or register to reply here.
Back
Top Bottom