The overdrive sounds good though. I have tried lots of bias voltages and found that under 82-83% sounds harsh.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
Try increasing the output valve grid-stopper resistors a lot - say 47k or so, even 100k, (rather than the 1.5k you usually see). You may find that you can then bias your tubes less hot without the harshness.
In my experience, the reason for the harshness is that, during overdrive, grid current starts to flow through the coupling caps that drive the output valve grids. This current flows in only one direction (grid-cathode acts as a diode), and that creates a DC voltage across those coupling caps, and that changes the bias on the output valves, pushing them colder and colder.
So the problem is that the valves are actually biased too cold, but only during overdrive, when grid current starts flowing.
The crude fix is to bias them super-hot, as you are doing - but that will shorten valve life, and limit clean headroom.
The more elegant fix is to reduce the amount of grid current that flows during overdrive - and that's what the 47k grid stoppers do.
If you're worried about the large grid stopper affecting the tone (removing treble), keep in mind that datasheet input capacitance for the EL84 is only about 11pF. That means a 47k grid stopper would only start to reduce treble above 309 kHz(!) It won't have any effect in the guitar's requency range.
In principle, you should make sure that total resistance from grid (G1) to cathode, including the new bigger-value grid stopper, doesn't exceed the datasheet maximum of 300k (fixed bias) or 1 megohm (cathode bias). In practice, guitar amps seem to routinely violate that particular specification.
By the way, so far I've found that I've never really enjoyed any amp with EL84 tubes. They always sound harsh to my ears.
Obviously, lots of guitarists love the sound of EL84s, so this is just a personal preference thing. But maybe your ears are like mine?
-Gnobuddy
I will try the larger grid stoppers. I have 10k in there now with 470k grid leak
As the amp is Single Ended I thought that the aim was to get 100% plate dissipation for it to sound best.
Sent from my iPhone using Tapatalk
As the amp is Single Ended I thought that the aim was to get 100% plate dissipation for it to sound best.
Sent from my iPhone using Tapatalk
Single-ended? You didn't mention that before, you know! I though we were talking about the usual push-pull class AB.
Single-ended changes things a bit. Yes, you now have to bias fairly hot, not necessarily for best sound, but if you want maximum power output.
Also, in class A (single-ended is class A), the bias current, B+ voltage, and OT primary impedance all decide when and how clipping will start - whether on positive peaks first, negative peaks first, or both at the same time. And that will affect overdrive timbre ("tone"), for sure.
The other part of the discussion - sliding bias due to grid current flow through coupling capacitors - is still true, both for single-ended and push-pull output stages. Hopefully, the larger grid stopper will help on that front.
-Gnobuddy
> thought that the aim was to get 100% plate dissipation
You also need to run a bit below 50% of maximum current at the voltages you run.
With low G2 voltage, the max (non-positive-grid) current is low.
You *also* want the tube V/I to be similar to load R.
At low voltages, "near max Pd" is often not a good goal.
At low voltages, try try try! You are not as likely to burn it up.
You also need to run a bit below 50% of maximum current at the voltages you run.
With low G2 voltage, the max (non-positive-grid) current is low.
You *also* want the tube V/I to be similar to load R.
At low voltages, "near max Pd" is often not a good goal.
At low voltages, try try try! You are not as likely to burn it up.
Today I dropped the el84 cathode bias resistor to 78 ohms ( parallel by 220uf)
Bias is now 4.85v and plate voltage is 213v. Screen is also 213v. HT / b+ from the rectifier is 232v .
So I calculate cathode current 62ma screen current is 4ma ( 2v dropped over 470 ohm ) plate current 58ma.
Plate dissipation is 12w (213-4.95)x 58ma
With 8ohm load on a 7k/16 ohm OT from an old radiogram load is 3K5 or approx equal to plate v on I as recommended.
The verdict it sounds great headroom is better to. Sounds quite good and less loud with 16 ohms.
Is that ok or by not running at lower current as you suggest PRR will I wear out the tubes quicker or cause some other sort of issue?
Thanks for the feedback
Sent from my iPhone using Tapatalk
Bias is now 4.85v and plate voltage is 213v. Screen is also 213v. HT / b+ from the rectifier is 232v .
So I calculate cathode current 62ma screen current is 4ma ( 2v dropped over 470 ohm ) plate current 58ma.
Plate dissipation is 12w (213-4.95)x 58ma
With 8ohm load on a 7k/16 ohm OT from an old radiogram load is 3K5 or approx equal to plate v on I as recommended.
The verdict it sounds great headroom is better to. Sounds quite good and less loud with 16 ohms.
Is that ok or by not running at lower current as you suggest PRR will I wear out the tubes quicker or cause some other sort of issue?
Thanks for the feedback
Sent from my iPhone using Tapatalk
- Status
- Not open for further replies.