The 10uf cap is definitely a lot better, so I will stick with that.
I tried to wire up the first half of the ecc83 to feed into the EF86, however, I get nothing but weird fuzz cutting in and out. I thought I had wired it up wrong so disconnected the EF86 and ran the ecc83 straight to the PI, which worked fine.
I'm beginning to think that the EF86 is good for just a stand alone tube rather than to daisy chain into other tubes.
I tried to wire up the first half of the ecc83 to feed into the EF86, however, I get nothing but weird fuzz cutting in and out. I thought I had wired it up wrong so disconnected the EF86 and ran the ecc83 straight to the PI, which worked fine.
I'm beginning to think that the EF86 is good for just a stand alone tube rather than to daisy chain into other tubes.
Another quick update. I tried running the second half of the ecc83 into the EF86 (which only has gain of about 4) and it worked fine.
So; I've got two ideas which I'm debating over:
The first plan is to run the ecc83 in parallel with gain of roughly 10 or;
Second plan is to reduce the gain of the first half of the ecc83 to about 10 and go along with my original idea.
So; I've got two ideas which I'm debating over:
The first plan is to run the ecc83 in parallel with gain of roughly 10 or;
Second plan is to reduce the gain of the first half of the ecc83 to about 10 and go along with my original idea.
I've never used an EF86, but have used other small-signal pentodes in my guitar preamp. They work well for me, as long as I am careful to keep them from generating too much voltage gain. The keys to this are to (a) lower the supply voltage, and (b) lower the anode resistor.
In addition to the gain issue, there is also the output impedance issue to deal with. A typical RCA/Fender 12AX7/ECC 83 stage with a 100k anode load will have an output impedance of around 40k. That's high, but at least it can drive something like a 500k volume pot without too much trouble. It will also generate a typical voltage gain of maybe x60 times, loaded.
On the other hand, that "classic" EF86 stage with an anode load of 220k will have an output impedance close to 220k. Very high! It will struggle even to drive a 500k pot. And the gain will be very high, as you said earlier, around 200 or more.
I run my small-signal pentodes on a B+ of around 150 volts, about half what a 12AX7 usually gets. I use anode loads in the ballpark of 40k or so, and then adjust screen voltage to get in the ballpark of 1 mA - 1.5 mA of anode current.
Now you have a reasonable output impedance (around 40k, like an ECC83 gain stage), and a reasonable voltage gain (around 60 - 70 for the pentodes I'm using, a gain that's also like an ECC83 gain stage).
Done that way, the pentode stages seem to be well-behaved, and more or less interchangeable with the more common 12AX7 stages you'll find in most classic guitar amp designs.
So far, I like having a pentode as the last stage of the preamp, rather than the first. In that position, it has some of the sonic character of a single-ended pentode power amp. I'm finding I can get some nice clean tones, and some nice overdriven tones too, out of the preamp. It sounds good to me, nothing like the rather harsh sounds of an overdriven 12AX7, which I don't care for too much.
-Gnobuddy
After doing some more testing, the only way I can get a decent enough sound out of the EF86 (with a preamp preceding it) is if I create a triode stage with a gain of roughly 2.4 (Anode = 100k, Cathode = 39k). So, My plan is to create a parallel ecc83 with a gain of roughly 2 (Anode = 10k, Cathode = 4.7k) and see if I can get this working.
If I had more time and facilities to test, I would love to play around a bit more with the EF86 and the screen voltages, but with basic bedroom equipment its a bit difficult.
I also have quite a few ECF80 valves which I think could be really interesting in a parallel combination.
If I had more time and facilities to test, I would love to play around a bit more with the EF86 and the screen voltages, but with basic bedroom equipment its a bit difficult.
I also have quite a few ECF80 valves which I think could be really interesting in a parallel combination.
There are going to be problems with such huge cathode resistors (39k) and such small anode resistors (10k).
The former will cause the stage to bias very "cold" (very little current). The latter will result in very limited output voltage swing.
In both cases, using local negative feedback to lower the gain so drastically, will also remove any trace of "valvey" sound; it will sound like a good solid-state Hi-Fi amp, because the feedback will make it extremely linear.
May I suggest reading (or re-reading) the chapter on gain stages at Merlin Blencowe's website? Here's a direct link: The Valve Wizard
-Gnobuddy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.