Bias for ECC99 Mini Vox AC30?

[roseblood11]

Hello,

I built a mini version of the Vox AC-30, but it doesn't sound exactly as it should. It's not as bright and "jangly" as a real Vox, and it might have too much bass.
My built looks like this:
Top Boost and Normal channels (wired to different sides of the PI, with sluckey's channel bridge switch at the inputs),
LTP phase inverter with Vox values,
150nF coupling caps,
250k stereo pot as PPIMV (instead of the 220k resistors),
22k grid stoppers,
one ECC99 as push/pull,
cathode biased with 330R and 47µF.
Output transformer is a Hammond 125C wired for 7k primary into a 10" 3.2 Ohm ELAC speaker (from an original 1965 AC10). I had the transformer wired for 15k before, but somebody suggested to go lower, maybe it sounds a little bit brighter now. The speaker is a good choice, I used a Weber Blue Pup before, which had way too much bass.

Plate voltage is 325V, cathode is at 10.6V.

According to Robs calculator, I get these values (see attachement). What would you suggest to get this amp closer to the real Vox sound?
I know that it will always sound "smaller" than an AC10 or even AC30, but at the moment it sounds much smaller and "boxier" than my DIY Vibro Champ (also with a 10" alnico speaker)

There are a layout and some pictures in this thread: https://musikding.rocks/wbb/index.php/Thread/419050-Mini-AC-30/


 Immo

[PRR]

> not as bright and "jangly" as a real Vox, and it might have too much bass.
> one ECC99 as push/pull

So you changed from pentodes to triodes?

Yeah, it will be "polite", not jangly. (Bass could go either way.)

Might be better with a pair of 6AU6 or similar small pentodes.

[kagliostro]

On other kit that uses 12AU7 or ECC99 the producer clearly say they didn't play as the original amp

but about the Molly they say it plays the same of the original AC30 (they say ..... may be)

On the link you can see the documentation as to compare your circuit with the Molly and see if they adopted some special trick to achieve the result

https://www.tube-town.net/cms/?DIY/LoW-Projekte/Molly

https://www.tube-town.net/cms/userfiles/media/low-molly/low-molly.pdf

https://www.tube-town.net/cms/userfiles/media/low-molly/low-molly-layout.pdf

Franco


p.s.:Their version use an EF86 .... don't remember if it was used in one of the early AC30, the EF86 claims for an AC15

[Greg]

Agree with both gents that it will not sound anything like the promptness of a pentode. So it's just the bass with no bad distortion or anything like that? To make your situation less bad would be to trade those 150nf for 22nf post inv. coupling caps... Sometimes cutting the bass is perceived as boosting treble, giving a clearer overall sound. Although I never tried it, the post inverter master volume is often associated with a certain boxyness in the taste critics worldwide...
In one of the links Franco provided, there's is a divider network cutting half the signal after the PI but that sort of thing can probably be accomplished with your PPIMV pot there.         

[Greg]

I often see instances of not going (also mini) in a pre-amp with the common idea to tame it down just before the power tubes with pots, R dividers network, series grid resistors and maybe lots of white noise? Maybe not...

I find that dividing circuitry (to tame gain and add tube saturation) can cut on the jangle in some cases. In EL-84 Voxes, there is plenty of jangle and the cut control might be handy. With small output triodes, I would think about not losing treble in the prior circuit as much as possible.

The top boost channel is a good idea, but I would try to replace the large tail phase inverter to a split load with unity gain (leaving some gain behind) and allowing you to use a whole 12ax7 section for make-up gain of your normal channel. Both channels will have the capacity to drive the output tubes, without the necessity to add dividers downstream.

The normal channel might be now slightly gainier than the top boost channel and can be set using common fender values 1k5/100k, and/or leaving the second k-cap unbypassed. Both volume controls could be bypassed with 200pf/500k or 100pf/1 meg using mica's. The top boost (since buffered) will still have plenty of what it is good for, with the gain almost as high as the normal channel and especially if you go 220k/2.2k on its first gain stage.

The first and second channels could enter in the cathodyne using 220k mixers (I don't think you'll ever need mixer resistors, both coupling caps channels (1nf) could act as mixers entering the cathodyne. The PI being about 15-20 meg input resistance, you will not lose treble there.

Both channels will be in phase for jumping them if you ever wished to do so. Could also replace the grid resistance of your ECC99's with 470k's without any master volume. That's how I would try it before scrapping it.

[tubeswell]

Some example load lines. If I was wanting to run this as a parallel output triode in single ended Class A operation with a 7k reflected load, I'd go for the lower (250V) plate voltage biased at -8.25V with a 200R cathode resistor. YMMV

[PRR]

250V at 20mA per side is 5 Watts. I'm seeing a 3.5W rating?
https://www.jj-electronic.com/images/stories/product/preamplifying_tubes/pdf/ecc99.pdf

.... ah, I see the calculator used by roseblood11 says 10W. Wonder who is right?

.... Someone I respect thinks at <=3W the useful life may be <1,000 hours, prompted by a guy who observed it becomes "grainy" (meaning it is distorted) in a 600 hour year.
https://www.diyaudio.com/forums/tubes-valves/205364-lifespan-jj-ecc99-tubes-post2879649.html
https://www.diyaudio.com/forums/tubes-valves/205364-lifespan-jj-ecc99-tubes-post2882272.html

This mirrors my experience in the 1970s working a 6BQ7 very hot. But those days 6BQ7 was going out of style and pretty cheap.

[roseblood11]

I played around with some values today and ended up with  270 Ohm + 22µF at the cathodes, 22nF coupling caps and 9k on the primary winding of the OT. And I cut out the cathode bypass cap of the second stage of the top boost channel (like it was in some 70s models).

Now it's really good for a bedroom amp. Still not as jangly as the AC30, but somehow like a hifi version of it. Less noisy and with a solid bass, because of a stiffer power supply and a diode rectifier, elevated heaters etc.

I thought about using pentodes, but I don't want to rebuild half of the amp. And the voltage is quite high for the mentioned small pentodes anyway.

I think I'll leave it as it is. :-)

Thanks for all the suggestions!

[tubeswell]

250V at 20mA per side is 5 Watts. I'm seeing a 3.5W rating?
https://www.jj-electronic.com/images/stories/product/preamplifying_tubes/pdf/ecc99.pdf

.... ah, I see the calculator used by roseblood11 says 10W. Wonder who is right?

I was going off the attached datasheet. (Maybe they've since de-rated the tube?)

[kagliostro]

I've always seen 5W per side as rating, 3.5W for the ecc99 was new to me

The thing is a bit .... confuse

https://www.jjtubes.eu/ecc99?search=ecc99

on the same page you can read 3.5W (gold pin version is the same of no gold pin) and on the same page on the pdf they say 5W

Franco