• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

newbie question

resident

Disabled Account
2004-09-10 4:57 pm
Earth
Hi,

How can I have more watts from this output stage without changing the anode voltage, that is 320V, and the primary of the OPT?
I don't want too many watts but as many as I can.

I will change it from UL to pentode but anything else? :confused:
I don't want to buy another PT.

The output tubes are EL34.

Thanks
 

Attachments

  • output stage el34.jpg
    output stage el34.jpg
    48.4 KB · Views: 240
You have half of it. Go pentode and convert to fixed bias. By going fixed bias the cathode resistors reduce in value and the voltage that was droped there is avalible for the plates. Just make sure the - bias voltage for the grids is very clean/well filtered/low noise and stable/regulated. I am sure you can find many circuits like this for the EL34 on the web.

BZ
 

resident

Disabled Account
2004-09-10 4:57 pm
Earth
I like the idea with fixed bias.I'll see if there is any space in the chassis for the needed tranny.
I'll see how much is the primary load of the opt,too.Maybe I can go with pentode.
Something like fixed bias with pentode.
Hope I'll have enough more watts.I'll check it.
Thanks guys!
 
If you go pentode, be sure to regulate the screen grid B+. You will get a lot less distortion to clean up via NFB.

FWIW, I think combination bias might be in order. Use unbypassed 100 Ohm cathode resistors and a bias supply. The slight increase in O/P impedance will not hurt, distortion is reduced, and the operating point stabilized.

You will have to increase the amount of loop NFB in order to get a damping factor roughly equivalent to that of UL operation. Fortunately, pentode mode has more gain to drive the NFB loop.

If you regulate the bias supply, you have to regulate the main B+ too. Either you regulate both the main B+ and the bias supply or you don't regulate at all.

Go deeper into Class "AB" operation to squeeze extra max. power out. Pick an idle current that's close to Class "B" operation, but not too close or you'll notice crossover (notching) distortion. 10 mA. of idle current might be worth trying.
 
Let me give you the bad news and the good news.

The bad news: the additional power you can get from that setup is minimal. Even if you double the power, that's only a 3dB difference, detectable but pretty minor. It really takes 10dB or so before you start really noticing "louder," all things being equal.

The good news: what you're really after is LOUDER. And you can do that relatively easily. The answer is to improve overload capability and recovery. An amp can be clipping away quite merrily without sounding horrible if the clipping doesn't cause squegging or other oscillations and you design the amp so that overload doesn't cause blocking. Taking overload into account can make an amp sound much cleaner at much louder levels.

In your output stage, you've got two sources of blocking, the cathode bypass caps and the grid input caps. By using, for example, a direct-coupled cathode follower to drive the output stage and provide negative grid bias, you can make your amp perform MUCH better when you crank it up.
 

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
> How can I have more watts from this output stage without changing the anode voltage.., and the primary of the OPT?

You can't. Power is V^2/R. If you don't change V, don't change R, how can you change Power? {oops: SY snuck in and said it first....}

OK, that's over-simplified. As others say, you can get a little more power by eliminating the UL connection. A pentode (often) lets the plate swing closer to cathode than a UL connection. Pentode may also be able to drive a lower load, but you say you don't want to change that. You can get a little more power by eliminating the waste in the cathode resistor and building a low-power grid bias supply.

Such changes give 10%-30% more power, maybe 50% more power if you do everything possible (including the odd inverted-UL or hyper-Pentode mode that dshortt9 suggests). 50% more power is 1.8dB louder. You can hear a 1dB change on a steady tone if you are listening for it; you can't hear a 2dB change in music if you don't know it is coming (and maybe if you do know it is coming).

"More Power", literally, and enough to notice, usually means significant change in supply voltage and/or load resistance (often enough to require a bigger tube).

"More power" can also mean "less strain at higher volume" or simply "louder". Sometimes more (or less) idle current sounds better. Sometimes a change in capacitors will make a tube amp sound like more power than before, because of less bias-shift on musical transients. Or in some situations, more bias-shift and rail-sag gives "impact" that a very-stable amp may seem to lack.

> I don't want to buy another PT.

What is your rectifier? If vacuum-rectifier, re-build it as silicon-rectifier. It may not be a lot more power, and some golden-ears say they don't sound right. I used to do that a lot, and get solid 50%+ increases in measured power. The amps also sounded better, to me, but I did a lot of other stuff in the re-building and that could mean more than the number of Watts.

> I don't want too many watts but as many as I can.

That's about a 20 Watt amp. Back in the old days, that was enough to make the neighbors complain. And they didn't complain any more about my 50 watt amps.

What speakers are you using? Back in those old days, we rarely went below 1% efficiency. But these days they sell little power-suckers as low as 0.2% efficiency. These make great sense on compact low-cost transistor amps: they are smaller than 1% efficient speakers. They don't make sense on tube amps: the amplifier ends up much bigger than the compact speaker.
 

resident

Disabled Account
2004-09-10 4:57 pm
Earth
As I understand the best solution is SY's suggestion.
But I can't change the driver stage.Everything is on a PCB so changes are not so easy.
That's why I decided to go with pentode and fixed bias.That's not too difficult.
I'd like to try pentode first.Just to see how it sounds.Maybe it's ok.
The primary load of the OPT is 3,3K.Is this good for pentode mode?Or to use the UL taps?If I'll use them then the primary load is lower than 3,3K so more power,right?
Before I go with fixed bias I can change the cathode resistors to go more to class AB right?Now I have 270R change them with what kind of value?
What is your rectifier? If vacuum-rectifier, re-build it as silicon-rectifier. It may not be a lot more power, and some golden-ears say they don't sound right. I used to do that a lot, and get solid 50%+ increases in measured power. The amps also sounded better, to me, but I did a lot of other stuff in the re-building and that could mean more than the number of Watts.
SS rectifier :(
That's about a 20 Watt amp. Back in the old days, that was enough to make the neighbors complain. And they didn't complain any more about my 50 watt amps.
I agree with you but this amp it's not mine.It's a friend's amp that he has around 86dB speakers and he is not happy with the power levels.
I own a 12W tube amp and with my Ariels MEII is just fine.
I don't know if I'll continue to hear my friend's complaints with his amp.I think that he must change speakers.But change it into pentode is not too much trouble.So I'll give it a try.
 
I don't think the changes your talking about here are going to help much. When you get done the small change in power will not make your friend happy because as SY said even if you double the power it's only 3 dB, not much of a change.

I would tell "my" friend to get some 94 dB/w or better speakers or a 100w/ channel amp.
 

ray_moth

Ex-Moderator
2004-01-27 8:55 am
Jakarta
Fixed bias is definitely worth doing, whether you stay with UL or not. You can then use the efficiency of AB1.

The primary load of the OPT is 3,3K.Is this good for pentode mode?Or to use the UL taps?If I'll use them then the primary load is lower than 3,3K so more power,right?

I very much doubt that the primary load is 3.3k for UL. With EL34s in UL PP, 6.6k is much more likely. You might like to check again.

For EL34s in pentode AB1 PP with fixed bias, you need about 3k to 3.5k plate-to-plate. Note that pentode mode is very critical concerning load: changing to a lower or higher impedance than the optimum will cause more distortion.

One way of changing the effective load would be to use different speaker taps. For instance, I have a 3.5k OPT with 4ohm and 8ohm secondary taps, indended for EL34s in pentode mode. I use them in triode mode, so I connect my 8 ohm speakers to the 4 ohm taps, effectively increasing the load impedance to 7k plate-to-plate, which is OK for triode mode. In your case, if it turns out that the existing p-p load impedance is actually 6.6k, as I suspect, you will want the opposite effect, so maybe you could try connecting the speakers to a higher impedance tap (e.g. 8 ohm speakers on 16 ohm tap). This is not an ideal solution but if you're stuck with the parts you have, then it might be worth trying.