Separate names with a comma.
Discussion in 'Shock Brother's DIY Amps' started by fensterman, Feb 23, 2017.
Here's some other suggestions for FX loops using 1 x 12AX7 in this situation. YMMV
I like the series/parallel loop
Here is the series/parallel loop
I saw a tweed bassman schematic with a loop installed.
I thought that looked interesting, I would need another preamp tube
With the virtual ground mixer in your schematic, the 1MA* master vol pot would form the load for the virtual ground, (as well as the grid leak path for the loop recovery stage) and you don't need the resistor labelled 'load'.
* although for this amp you could happy get away with a 250kA or 500kA pot for the MV, and it would be better for the grid leak resistance path.
Thanks , I was going to ask how to calculate the resistor value
That is what I was referring to in post 19. The loop in most useful after gain and EQ as shown. In your case: Gain>vol>gain>EQ>Snd>RTN>KT88. It's still 2 tubes. Or follow the Bassman schematic with the cathode follower, which is also 2 tubes.
I took the front end from a 5f6a. Let me know if you see any issues.
Yep, that's the idea methinks
I like it too, but the return side of the FX loop going into a driver tube is a lot of gain without having to throw most of it away. In fact, the SND and RTN sections can both be cathode followers, and the driver stage will still have plenty amplitude to drive the output.
When you start to play it, have a SS rectifier plug handy to swap out with the tube rectifier. It'll give the output tube a little bump, and the transformer will run a little cooler not having to provide 10 watts to keep the rectifier's heater warm.
The advantages of a virtual earth mixer for the FX loop recovery stage in this situation are its low input sensitivity and near unity gain, without needing to think about high h-k voltage. It can take whatever signal is being inputted into it from earlier in the preamp and spit this through without changing/distorting it dramatically. There will be ample gain in this amp with the '5F6A' front end. The FX unit should be able to supply enough line level signal to feed through into the driver stage, and there shouldn't need to be any extra gain needed.
That's a better idea, leave the RTN circuit alone, and loose the following driver stage.
If I remove the driver tube I can put both inputs in. Then I can copy the front end of a 5f6a board
In that last schematic you posted, you also need a method for biasing the cathode follower stage that you are using to drive the FX send. You have the CF stage AC-coupled to the tone stack, so you either need something like a voltage divider for the grid bias voltage (like the bypass stage in a 6G15), or cathode-biasing with a grid leak resistor bootstrapping the gird to the junction of the load resistor and a bias resistor.
How about this method?
I think there will be way too much unusable gain there. I would say you'd be better off driving the FX send from a split load on the CF driving the tone stack, and having a more neutral FX return gain structure (like a virtual earth mixer), with a switchable series parallel loop (like the FX return setup you had previously. YMMV
Okay, lets backup to version 25.
First, you have to educate me here.
We will try to follow along to fine tune that schematic.
High h-k voltage?
Ok , Maybe we can make some progress.
I removed the driver and added a voltage divider on the FX send section.
'h-k voltage' = 'heater-to-cathode voltage'.
In a cathode follower, the cathode typically sits at (say) 100-200VDC at idle. The heater is at 6.3VAC (which is 8.9Vp-p). This leaves (in our example) 90 to 190V potential between the cathode and the heater filament (which are physically positioned right next to each other in the electrode cage). The heater-to-cathode insulation ('h-k insulation') that is squirted onto the filament during production, is rated for a certain potential (in a 12AX7, this is 180V, in a 12AU7 it is 90V, on most data sheets). The higher the h-k voltage, the more likely that the h-k insulation will fail. This is why modern production* pre-amp tubes suffer in high-gain amps where there are a lot of CF stages.
* In the good old days where everyone made tubes and were concerned that only the highest standard of tubes should leave the factory, tubes that made it to the retail shelves could take a beating better, than what is available today.
Ideally, an FX loop should provide a clean line level signal on the send jack, which is more than sufficient to drive FX units, and a recovery stage at the return jack that can process the returning signal undistorted and mix it back in to the amp's signal path.
A CF stage has unity gain and capability to take an input signal with rail-to-rail signal amplitude and spit this out without distortion, so is an excellent candidate for an FX return stage (and if the CF load resistor is sufficiently split to enable line-level signal output that won't overdrive and FX unit input, can also be a suitable type of stage for driving an FX send). However CF stages exhibit a high h-k potential difference, and typically require substantial heater elevation to overcome the tendency for h-k insulation failure.
In comparison, a virtual earth mixing stage has almost-unity gain, so it is a good candidate for taking whatever signal is being spat out of an FX unit, without adding extra distortion, and feeding this back into the amp's signal path. The advantage of a virtual earth mixing stage is it doesn't require high h-k operating voltage (and even though it doesn't quite provide unity gain, it is 'close enough' to make a good FX return stage), and so its probably gonna work better with modern production tubes.
That's my preference anyhow. But you may well like something different.
Any comments on the last schematic?