Altec 1568A Proper Bias Level

[Bryrohr]

Hello,

I recently purchased a pair of Altec 1568A EL34 based mono tube amps. They work excellent and I plan on keeping them all original as they just have too much vintage appeal to change them. One thing I have noticed is there are two versions of factory released schematics running around the web that appear to be identical except for one has a -30 volt bias level and the other has a -37 volt bias level. Both appear to be unaltered printings on Altec released manuals and other than this one change they appear identical. The amps I have were set at very close to -30 volts bias and operate fine at this setting and sound absolutely astounding. I have have many other power amps that one would think would simply blow these out of the water and I really didn't expect much with these and was pleasantly surprised. The only thing I can see would happen by changing the setting to -37 volts would do would decrease the idle current through the EL34s and probably increase tube life, however, would lowering the plate current in this manor change the sound that much as well?

Any comments would greatly be appreciated.

Thanks

[tube fanatic]

Didn't have time to study the schematics, but a quick glance suggests that your interpretation is correct. By lowering the current a bit, the plate/screen voltages will run a bit higher. This, of course, will affect the required load impedance somewhat. Since Altec used the same output xfmr in both units, it apparently isn't of much concern. But, if all is well with the -30 volt bias, why mess with it? You could always put a 'scope on the output (with a proper load connected) and see what effect the altered bias has..................

[Bryrohr]

Fanatic,

I do have a scope I could look at the change with. I assume
doing this with a dummy load (I also have) at or near the maximum rated output
level of 40 watts at say 1kHz would probably be a good place to start? I'm
guessing the bias level shouldn't have an extreme difference on the output
capacity since the B+ will stay the same roughly and the output
devices/trannies are the same as well, but if the bias is set radically too low the tube would
idle too hot and power output would decrease significantly as would tube life.
Is this a good assumption?

B.

[tube fanatic]

Sounds like you want to have some "fun" with this! By chance, is your 'scope a dual trace type with which you can superimpose the waveforms? That would make it easier to compare your sine/square wave input with what is coming out. But, in any event, you can connect your 8 or 4 ohm resistive load (assuming it's rated for the power needed) and look at the sine wave appearance at the point just before the onset of visible clipping (a flattening of either the top or bottom rounded portion of the waveform. In most amps, the clipping will not be symmetrical). Of course, don't leave it running like that for too long. Compare that to the appearance of the input signal from your generator. There are some nice articles online which show how to interpret deviations in the waveform appearance and form some correlation with distortion. You can try that for multiple frequencies to get some appreciation of what's happening across the audio band. Do the same with a square wave input to see how things behave with a more "complex" signal that effectively encompasses a range from around 1/10 the input freq. to 10X the input frequency. With this, however, I wouldn't go beyond about 1 watt output (keep your multimeter, set to an AC range, across the dummy load to monitor the voltage- keep it to 2.83 V maximum). Again, check out the online articles on that- it's much easier to explain with pix than in words. I suspect that you're not going to see a huge difference. Considering that tube characteristics can change quite a bit as they age, a 7 volt bias variation could occur with normal use over time. Most users of tube equipment, unless they're really into the technical side, never check tube bias anyway. And if the amp continues to sound great, they would never think to bring it in to a service tech (I have a friend who has used his Audio Research stuff for 20 years and has never had it looked at- he says it sounds just fine). And, yes, if you allow the tubes to draw too much current by biasing very low, life span will definitely be reduced as you might exceed the maximum plate/screen dissipation rating of the tubes. So, I hope this helps you a little bit. Post or PM if I can help any further.

[Bryrohr]

As a matter of fact I do have a dual trace (rather slow at 15Mhz but not a
problem for audio stuff) scope, also have two 8 ohm @ 250 watt Dale power
resistors, signal generator for both sine and square, variac, Audio Control
SA3050 RTA, various Fluke multi-meters...Etc.

I agree with the low signal when feeding the input square waves as I
wouldn't want to fry the trannies or pwr sply. Tubes are all NOS and barely
broken in with maybe 50 hours on them.

However, I did feel the 7 volt deviation of bias spec could matter as this
would represent nearly 25% of -30 volts. I understand that if I were to run the bias
voltage to -37 volts it would lower the plate current but what effect on sound quality
and output capacity would result in this change is a mystery to me.

My basic knowledge tells me that if the amp performs well, output tubes arent running at
too high a plate current and the amp sounds great at the higher (-30vdc bias)
plate current, go with that. My major concern is why Altec printed two
schematics with two very different bias levels and what they thought was correct. Do
you think they later changed the original bias spec of -30 vdc to -37 in an attempt to make the amp more
bullet-proof and have longer tube life, or do you think they first chose -37
volts and then they felt the plate current just needed to be higher for better operation and
sound quality so they revised it to -30 volts? Honestly, I haven’t even checked
to see if the bias supply circuit is capable of producing -37 volts DC. LOL. If it won’t,
I can rest assured this is just a typo on Altecs behalf.

I'm thinking I should just measure the plate current at either setting and
see what value fits the EL34's recommended plate current curves the best.

[tube fanatic]

The difference in bias voltage between the 2 schematics remains a mystery. It would have been helpful if they printed a serial # range for each one, and even more helpful if they specified the plate voltages of the el34s. It certainly is possible that they decided, at a later date, to run the tubes cooler to increase longevity. As far as sound quality with the two bias voltages, why not reset it to -37 and take a listen? You certainly won't hurt anything by running the tubes at a lower plate current. The chances are, even if it does increase the distortion a bit, you may not even notice a difference when playing music. As far as output power, measure it at 1 kHz at the onset of clipping on the 'scope and compare with the two bias values. It might be a fun experiment! And you can certainly measure the plate current/voltage/screen voltage at each bias value, and graph it at the two bias voltages (I still do it the old fashioned way on paper, with a ruler, etc. If you know someone with a CAD program it sure will save you a bunch of time). You would then need to measure the transformer impedance (easy to do with your signal generator) to see which provides the closest match. Who knows, you may find that the "correct" bias voltage based on the graph is in between the two values printed! If you have, or can borrow, a copy of the Radiotron Designer's Handbook it will give you everything you need to graph virtually anything. It's been my "bible" for longer than I can remember. In any event, you certainly have a good handle on what's needed here to answer the bias question. Also, in case you are not familiar with it, check out www.tubedata.org which has the most extensive database of tube characteristics anywhere in the world. Please post your results after doing the experimentation. I have lived and breathed tubes for the past 50 years and still marvel at the amazing things that they can do! Your last name is definitely a credit to your interest in the subject!!!