Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Leevi on September 07, 2012, 02:36:24 pm
-
I have recently built two amps that are based on Fender 5F2-A circuit (Princeton). Both are without NFB.
I used in the first amp 6V6 power tube like Fender did but in the second amp I changed the 6V6 to EL84.
The amps sound totally different.
The 6V6 amp has lot of headroom but the EL84 version starts to overdrivealready from 8 o'clock position.
Any idea what could be the reason for this? The preamps are equal like the PT.
/Leevi
-
Bias are not too hot for the EL84 ? EL84 and 6V6 have the same power but this do not mean you can put any of those without checking bias .
Another tube = bias check . Same if you replace a 6V6 by another 6V6 , class A or AB amp.
-
AFAIK... the EL84 is much easier to drive than the 6V6 which is why the EL84 amp seems overly gainy.
-
Yes the EL84 has a typically lower bias voltage than a 6V6, so it goes into grid clipping 'sooner' than a 6V6
-
Hi Leevi
you must adjust drive signal level for el84
try with a voltage divider between driver and final tube
or you must play with anode and/or cathode values of V1b
or a voltage divider between V1a and V1b
K
-
Bias are not too hot for the EL84 ? EL84 and 6V6 have the same power but this do not mean you can put any of those without checking bias
I forgot mention that I set the bias separately for EL84 so that the idle cathode current is about 48mA -> 10V.
you must adjust drive signal level for el84
Yes, that would be the corrective action. I just wondered the difference in the power tubes.
Leevi
-
Bias are not too hot for the EL84 ? EL84 and 6V6 have the same power but this do not mean you can put any of those without checking bias
I forgot mention that I set the bias separately for EL84 so that the idle cathode current is about 48mA -> 10V.
you must adjust drive signal level for el84
Yes, that would be the corrective action. I just wondered the difference in the power tubes.
Leevi
The 10V grid-to-cathode voltage means you have 20V-PP before clipping starts at the grid
Whereas a typical 6V6 bias voltage is about 20V grid-to-cathode (hence the later break up for this tube)
-
20V-PP
tubeswell, what do you mean by that (Push Pull?).
Whereas a typical 6V6 bias voltage is about 20V grid-to-cathode (hence the later break up for this tube)
Should I increase the value of the cathode resistor in order to get the voltage to 20V?
/Leevi
-
Whereas a typical 6V6 bias voltage is about 20V grid-to-cathode (hence the later break up for this tube)
Should I increase the value of the cathode resistor in order to get the voltage to 20V?
/Leevi
No, not for the EL84 amp.
Do yourself a favor: open both versions of this amp and measure the bias voltage of each. You will find the EL84 amp has a smaller bias voltage.
The EL84 has higher Gm than the 6V6. Gm (transconductance) is defined as a small voltage change at the grid causing a resulting plate current change, and in Europe is written in units of mA/V. Because the EL84's Gm is higher, a smaller grid voltage change results in a larger plate current change than in the 6V6.
The result is that it take a smaller change in bias voltage to turn down the EL84's plate current, and the bias voltage is smaller than for the 6V6. This is very normal, and you should not try to apply the 6V6's typical bias voltage to the EL84, or it will be be pushed further towards cut-off.
On the flip-side, as said, the EL84 now needs a smaller signal to drive it to full output power. This is a good thing, and the high Gm of the EL84, EL34 and EL37 is what makes these tubes attractive. They are easier to drive than other comparable output tubes. Again, think in terms of a needed resulting plate current swing, and mA/V. Higher Gm means it takes fewer V's to get all your mA's.
The correct approach is to reduce the gain of the preamp some how. If you want both amps to respond the same way at the same volume control setting, and don't want to make changes to tube stages which might lead to other side-effects, the alter the volume control. You can use a resistor between the ungrounded outside end of the volume control and the source of its signal. This makes it so the voltage divider created by the volume control can never pass 100% of its input voltage.
If your volume control is 1MΩ, you might want to try a 250k or 500k pot wired as a rheostat to determine the resistance that gives the desired volume control operation. Measure the final resistance represented by this rheostat, and replace it with a fixed resistor.
-
Thanks HotBluePlates, this (Gm) clarifies the difference.
I found a transconductance value for EL84 which is 11.3mA/V ("mutual").
For 6V6 the values are given with umhos e.g. 3700 umhos. Probably there is a converter somewhere?
I already added a voltage divider to preamp and dropped the signal level little bit on the power tube grid.
These changes together ganged the situation a lot.
/Leevi
-
20V-PP
tubeswell, what do you mean by that (Push Pull?).
peak-to-peak
Whereas a typical 6V6 bias voltage is about 20V grid-to-cathode (hence the later break up for this tube)
Should I increase the value of the cathode resistor in order to get the voltage to 20V?
/Leevi
No. An EL84 doesn't need a higher bias voltage to work efficiently because it has higher transconductance than a 6V6. If you increase the bias voltage to 20V on an EL84 it will be a very sad and cold tube (if it indeed is still conducting at that voltage)
-
I found a transconductance value for EL84 which is 11.3mA/V ("mutual").
For 6V6 the values are given with umhos e.g. 3700 umhos. Probably there is a converter somewhere?
I'll be your converter. 3700 micromhos = 3.7mA/V. And 11.3mA/V = 11 300 micromhos.
Ohm (R) = Voltage/Current = V/I
Mho = 1/ohm = Current/Voltage = I/V
1 millimho = 1mA/V = 1000 micromhos
-
I'll be your converter. 3700 micromhos = 3.7mA/V. And 11.3mA/V = 11 300 micromhos.
So the difference is huge.
Thanks
/Leevi