EL84 Voltage/current readings

[charleyrose]

Hi,
I would like to post my evaluation of the following new production
EL84 tubes as installed in my ST-35 "original" dynaco amplifier.
Unit is completely stock (recently restored) with the exception of
a 100 ohm 5 watt fixed bias resistor.
All tubes tested a new "matched quads"
(a) EI Gold series total current 153 ma (dc) @ 15.3 vdc
EI std series total current 150 ma (dc) @ 15.0 vdc
JJ brand total current 147 ma (dc) @ 14.7 vdc
Sovtek brand total current 125 ma (dc) @ 12.5 vdc

Please understand I am just a beginner with vacuum tubes with
minimal electrical background. I do respect the lethal voltages.

Based on the above readings, it appears to me that the JJ brand
tubes are the nearest to the recommended 140 ma total current
measurement, although a bit high at (+7 ma).
As suggested by Ed B (NH) the bias resistor could be increased to
obtain the desired 140 ma total current or alternatively a 15 ohm
5 watt potentiomter could be added in series with the existing 100 ohm
resistor.
Based on the Sovtek readings it appears that I would need to go the
other way reducing the value of the bias resistor. Correct ?
Obviously, the potentiometer option would allow for use of almost any
brand tubes and this is under consideration although I am not 100%
sure how to wire this correctly. I have wired simple volume controls
before and assume that it is basically the same routine using a linear
pot with the (3rd) leg to ground ? Is this connection needed as I believe
the 100 ohm resistor is already grounded ?
A prior post stated that new production EL84 tubes tend to run hotter as
compared to earlier production tubes such as "Mullard".
Does this imply that the newer production tubes tend to draw a higher
current and the 140 ma recommended current spec is not that critical ?
As an an example, the JJ band tubes were running at a total current of
147 ma. Would this be considered to be in the prefered current range.
On the "flip side" side of this would running the tubes below the 140 ma
current as is the situation with the Sovtek brand tubes be adverse to
the operation of the amplifier and/or sound quality.
Sorry for all these questions; I look forward to contributing to this
Forum in the future. Thanks to all Charley Rose

[erichayes]

Hi All,

Charley, you have stumbled into the wonderful/horrible world of Cathode Bias, also known as Self Bias. Perhaps if I flesh out the bias systems used you'll be able to answer some of your own questions. Let's start with fixed bias.

Fixed bias applies an unvarying negative voltage between the grid and cathode, with the cathode at or very near ground potential (0vdc). The bias voltage can be adjusted so that the grid never goes positive on signal peaks (class A1) to a point where the grid draws some current, but not exceeding the tubes' spec (class AB2). The class of operation can easily be changed by simply adjusting the bias control for whatever cathode current is spec'd for the particular tube. The bias voltage doesn't need to be regulated, as no (or very little) current is being drawn, but it needs to be constant to avoid distortion. When running in class AB1 or AB2, the cathode current will vary in direct proportion to how hard the amp is working--moreso with AB2.

Cathode bias, on the other hand, puts the grid at ground potential and raises the cathode off of ground by putting a resistor, usually between 100 and 500Ω, between the cathode and ground. The grid is now negative with respect to the cathode, which is all we're really concerned with. The method is simple and cheap (which is why it was developed), and, for the most part, is perfectly acceptable sonically.

There are a couple of downside issues with cathode bias: they tend to have lower output than their fixed bias equivalents, and they tend not to have the dynamic range and solid low end. This is because the cathode resistor acts like a ballast and absorbs the current peaks to some extent.

6BQ5s almost always used 250Ω resistors on their cathodes, or 130Ω per pair in push-pull. Dynaco went with the equivalent of a 380Ω resistor per cathode, but ran the plate voltage higher to get the output power where they wanted it. The problem with this approach--putting all the cathodes on one resistor-- is that you're condemning all your output tubes to death if only one fails to draw its fair share of current due to heater burnout, bad socket connection, etc, because the other tube(s) are now connected to a resistor that wants to see the current of, in the ST-35's case, all four tubes. That means the remaining 3 are going to hava to take up the slack.

Output tubes, in general, and especially in cathode bias operation, are pretty tolerant of deviations from their design parameters. The numbers you got on your tubes are well within the tolerance of the 6BQ5 specs. If you want to experiment with how the amp sounds using pots or different value fixed resistors, this amp provides a good test bed.

[dhuebert]

I have an AGS stereo receiver which has a quartet of 6BM8s on the output (two per channel). All four are on the same cathode resistor, guess what - every time I turn it on a different tube glows red hot! If this were the only problem I would fix it and put the thing into service but it is the worst designed (mechanical as well as electrical) piece of crap I have ever seen, so it goes into the parts bin.

Don

[Shannon Parks]

I have an AGS stereo receiver which has a quartet of 6BM8s on the output (two per channel). All four are on the same cathode resistor, guess what - every time I turn it on a different tube glows red hot!

In my hurry to finish up my Eico HF86 restoration several weeks back, I left out two grid resistors - one on each pair of the output tubes. DOH! Two of the EL84's ran hot (though not thermal meltdown) while the other two didn't. The shared cathode resistors threw me off a little. But I should have ohm'd it out first. Yellow_Light_Colorz_PDT_14

I would do a simple 'ohm out' test on this one, Don. Could be a couple small problems. See that pin 1 matches on all four, then pin 2, etc.

Shannon