idle current vs breakup - any correlation?

[bluesholyman]

I am trying to understand if there is a correlation between a power tubes idle current, 6V6 specifically, and its breakup point.

Does a lower idle current mean a faster breakup, or does a higher idle current mean a faster breakup?

Or do I not understand and should just watch the pretty glow...?

[Stevem]

Higher current= faster break up and depending on how the tube handles being overdrivin looser bass / low end, but this is very dependant on how much of the tubes wattage output the amp is set up for.
I find most 6v6s other than JJs like about .018 to .022 amps which again is hinged to how much B+ you are hitting them with and the OTs primary resistance.
If you go to the Duncan amps tube site and use the tube data locator you can see the specs, but down on the bottom of that page click on the specs from triode electronics for more detailed info.

[martin manning]

I am trying to understand if there is a correlation between a power tubes idle current, 6V6 specifically, and its breakup point.

Does a lower idle current mean a faster breakup, or does a higher idle current mean a faster breakup?

What kind of power section? A single ended amp could go either way with higher idle current depending on how well centered the operating point is, while a push-pull arrangement will most likely break-up sooner because the grid voltage between the bias point and zero is being reduced.

[Leo_Gnardo]

Does a lower idle current mean a faster breakup, or does a higher idle current mean a faster breakup?

Insufficient bias current will have your push pull amp sounding distorted all the time. Harry Kolbe, famous NYC amp goo roo used to set bias @ 12 to 15 milliamps on Marshall EL34's because "you rock guitarists just want distortion all the time." Well for the few that do, I s'pose that's a solution.

Single ended? Insufficient bias current will sound thin and weedy. Maybe some like that too, but it's sure not for everybody. Easy enough to experiment for instance with a Champ or similar amp, just try a range of cathode resistors and your ears will tell you what's what.

Excess bias current will sound poorly defined in the bass, and early breakup due to reasons Martin mentioned already.

[teemuk]

http://diyaudioprojects.com/mirror/memb ... /pent.html

The whole page is worth reading - several times actually - but you are probably most interested in section titled "The Effect of changing the bias on Pentode Composite Curves".

[tubeswell]

Tube idle current is related to bias (grid-to-cathode) voltage and plate voltage and (in a tetrode or pentode) screen voltage.

(Leaving screen voltage aside for a moment,) the higher the plate voltage, the higher the tube idle current will be - all other things being equal. And, the lower the bias voltage, the higher the tube idle current will be - all other things being equal. So if we want to keep the plate happy, we increase the (negative) bias voltage slightly - if the plate voltage is higher - in order to keep the plate current from getting too high (or conversely, to keep the screen current from getting too high when the plate voltage is high - which happens during the 'low-plate-current' part of the tube's signal swing).

But all other things usually aren't equal, because when you increase the tube idle current by lowering the bias voltage, you will invariably lower the plate (and screen) voltage, because the higher current 'pulls' the B+ voltage down because there is more current being drawn though the HT winding, which puts more load on the power transformer. Too much more current can make the plate (and screen) heat up in a way that turns the blue glow into a red glow - so watch your plate current x plate voltage.

Clipping occurs at the tube's control (signal) grid when the peak signal swing at the grid exceeds the bias (grid-to-cathode) voltage (and remember, the bigger the bias voltage, the lower the idle current). If the grid signal peak is 'too negative', the tube will be driven into cutoff (because the grid voltage at that point will be so low that tube (plate and screen) current is effectively turned off). If the grid signal swing is 'too positive' (i.e.; if the grid signal swing positive peak voltage approaches zero grid-to-cathode volts - remember that we have set the bias (grid to cathode) voltage negative at idle, in order to stop the plate and screen current getting out of control -), then the grid signal will get clipped on the positive side, because the grid signal voltage can't go into absolute positive territory due to grid-current limiting.

Now back to the screen voltage. If you increase the screen voltage, this increases the amount of change in plate current for a given amount of change in control grid voltage (i.e.; during the signal swing). This is because the screen is effectively an electron particle accelerator, and acts to pull more electrons across the gap between the cathode and the plate - so more screen voltage encourages relatively more electrons to flow when the control grid voltage swings more positive. So this is why we see the space between the control grid curves increase when the screen voltage is increased, and why screen voltage affects tube bias (all other things being equal)

One more thing related to distortion - more much plate current translates into more magnetising current through the OT core (when the signal is going through the OT). If the OT core is smaller, this can cause saturation of the core at lower frequencies, resulting loss of definition of bass bandwidth.

[talbany]

Clipping occurs at the tube's control (signal) grid when the peak signal swing at the grid exceeds the bias (grid-to-cathode) voltage (and remember, the bigger the bias voltage, the lower the idle current). If the grid signal peak is 'too negative', the tube will be driven into cutoff (because the grid voltage at that point will be so low that tube (plate and screen) current is effectively turned off). If the grid signal swing is 'too positive' (i.e.; if the grid signal swing positive peak voltage approaches zero grid-to-cathode volts - remember that we have set the bias (grid to cathode) voltage negative at idle, in order to stop the plate and screen current getting out of control -), then the grid signal will get clipped on the positive side, because the grid signal voltage can't go into absolute positive territory due to grid-current limiting.

One other thing to add so we don't confuse clamping with clipping!

Grid waveform clipping usually occurs at the PI. The grid waveform clamping is ocurring at the grid ,(however, the clipping can be caused by the clamping as well since it's now acting as a forward bias diode) If the output is clipped, the input will have hit the grid rail, so it is just a matter of whether or not the signal is AC coupled and clamped or not.

The way to spot grid clamping is to stick a scope probe (DC coupled) on the grid while slowly cranking up the level. You'll see a point where the top of the grid waveform will stop moving up , but the bottom will keep moving down. That's grid clamping. At some point, the top and bottom will clip.

Several way's to relieve or eliminate the effects of grid clamping are?..Speed up the bias excursion recovery time, or simply source and sink grid current by adding a low impedance DC coupled CF driver stage. Basically going into AB2



Tony

[pdf64]

Tony, surely for an AB1 amp, signal clamping at the power tube control grid equates to the max signal that the amp can put out, so control grid clamping = power amp clipping?

As you point out, the only way to move beyond that is to change the class of operation to AB2.
Pete