Taming amp gain in single channel deluxe reverb build

[imo1]

It does sound nice with a 335! I have a ‘65 335 and a ‘55 Jr. Both sounded fantastic!
The pre PI gain attenuation really is huge. I messed around with dumping gain in a couple of other places, but that is really the money spot.
I also knocked the cathode bypass down to 5uF on the first two stages.
At a certain point it all comes down to the player, but to me this solves all the deluxe “problems” while retaining what I love about them. It still has that softness and clarity and blooms at a totally usable volume.

[lonote]

Wondering if you have an updated final schematic?

I have a similar amp nearing completion, seems to have more gain than I expected or want. I would be interested to see where you ended up.

[martin manning]

I was thinking the same thing... I read through the thread and tried to capture the changes in the attached.

(updated)

[imo1]

Wondering if you have an updated final schematic?

I have a similar amp nearing completion, seems to have more gain than I expected or want. I would be interested to see where you ended up.

I didn’t do a schematic with my changes. I did, however, go through the amp following the instructions fender tech notes with expected voltages. What I found is that the design I built maintained a general linearity similar to a production deluxe reverb until around the phase inverter.

The 47k to ground before the 220k mixing resistor is supposed to account for the absence of the tremolo circuit but I still had way to much gain. It was suggested to try a 250k pot in place of the 330k. That was really helpful and I found that it really is a crucial component of the sound. I ended up picking an 82k for that spot and with that resistor there, the quality of the amp stays pretty clean, even on 10. If you go to a 100k or a little more, it starts to sound a little more like a typical deluxe towards the top of the volume.

I used a SS rectifier circuit so I also was able to double the first node to 94uF. I kept the other nodes as they were(22uF) That tightens up the woofiness some, and finally, I changed the first 2 cathode caps to 5uF, which, again, helps to tighten up the bass.

I used a 1M pot for the reverb dwell. The sweep of the pot isn’t perfect, but the effect worked for what I needed.



These are all taste things at a point, but I’ve owned and played a number of deluxes and other mid volume fender amps and always struggled with the low end farting out, so to me, these are major improvements

IMG_2576.jpeg

[imo1]

I was thinking the same thing... I read through the thread and tried to capture the changes in the attached.

Questions:
Are the other filters 16u or 22u?
Is the Reverb Dwell a 1M linear pot?

Martin,

You posted just as I was. Yes. Those changes are all the ones I implemented. Thanks for the markup. The 33r I forgot to note in my reply, but that was accounting for the tap taken.

[martin manning]

Ok, cool. I'll fix up the schematic based on the above.

[martin manning]

Those changes are all the ones I implemented...

Fixed a couple of things in the mark-up. Where did B+ land? Any changes to the dropping string?

[lonote]

Thank you both for the info & the markup.


Looks like I need to fool with PI input voltage divider as implemented here.


I have built nearly this exact circuit, including the dwell, SS rectifier & choke, with the exception of using EL84s & a few different resistor values in the PI because of the different power tubes.

[Roe]

add grid resistors downstream and reduce the values of the coupling caps to reduce low bass, esp. early in the circuit. see if the filtering is sufficient

[imo1]

Those changes are all the ones I implemented...

Fixed a couple of things in the mark-up. Where did B+ land? Any changes to the dropping string?

I don’t have my notes in front of me(on tour) but the 6v6s were right at 420 and all of the preamps were pretty much right in spec with the deluxe schematic. So thinking 185-190 on v1, etc.

Nothing much higher or lower than spec.

My bias was taken off the hv winding and I implemented Merlin’s bias circuit, but used 10k pot and resistor.

[imo1]

add grid resistors downstream and reduce the values of the coupling caps to reduce low bass, esp. early in the circuit. see if the filtering is sufficient

I messed a little bit with coupling caps, but preferred what the smaller cathode caps did in this circuit. There is a nice clear bass that I can turn up to 10, but because of the increased filtering and cathode caps, it holds together.

If I play on 10 on my neck pickup with the bass way up it starts to get clubby, but that’s an extreme case of usability

[Roe]

add grid resistors downstream and reduce the values of the coupling caps to reduce low bass, esp. early in the circuit. see if the filtering is sufficient

I messed a little bit with coupling caps, but preferred what the smaller cathode caps did in this circuit. There is a nice clear bass that I can turn up to 10, but because of the increased filtering and cathode caps, it holds together.

If I play on 10 on my neck pickup with the bass way up it starts to get clubby, but that’s an extreme case of usability

good! grid resistors also help minimizing blocking distortion. and be careful with .1uf coupling caps in the PI combined with 220k bias splitters

[martin manning]

I don’t have my notes in front of me(on tour) but the 6v6s were right at 420 and all of the preamps were pretty much right in spec with the deluxe schematic. So thinking 185-190 on v1, etc.
Nothing much higher or lower than spec.
My bias was taken off the hv winding and I implemented Merlin’s bias circuit, but used 10k pot and resistor.

Ok, thanks. I made another revision, with suggested current production Hammond power transformers which include the bias tap, and left the original DR bias supply. I would have thought that without the normal channel 12AX7 you'd have a bit higher voltages on the V1 plates at the same 6V6 plate voltage.

[lonote]

grid resistors also help minimizing blocking distortion. and be careful with .1uf coupling caps in the PI combined with 220k bias splitters

On my project, I was able to experiment some yesterday with knocking back the PI input using a voltage divider (the 220K + an added 250K pot to ground) & made some progress. Somewhere around 40K-60K on the pot was working well.

The comment about the .1uF coupling caps stands out though, as I have both the .1's & the 220K bias feeds in this circuit. Not sure how I landed on that CC value, other than auto-pilot.

As I am finding this one too bassy & will sub in some different CC's next.

Seems like integrating the Fender preamp to the EL84s is more fussy that I expected.

[imo1]

I don’t have the specs but don’t el84s need a good deal less gain to drive them then 6v6s? That would make sense.