Effects Loop in simplified AB763 Deluxe Reverb

[Spectre]

Hi,

I’m designing/building a simplified (single channel, no reverb, no vibrato) AB763 Deluxe Reverb based around Rob Robinette’s Blackvibe.

I’m looking to incorporate Merlin Blencowe’s simple serial effects loop.

See the layout attached. I’ve “broken” the signal before the 0.001uF Cap that is on the first grid of the PI. It returns at this same Cap.

I’d appreciate any comments/advice/other on the design, component values etc from those with better technical knowledge and experience as I’m a still a bit of a novice when it comes to this stuff.

Thanks very much.

[pdf64]

The layout picture looks lovely, but to comment on the circuit design, a schematic is required

[Ten Over]

The signal at the cathode of the FX Send triode is around half the signal at the grid. Between this and the cascaded voltage dividers, the signal at the FX Send will be too small at moderate settings. The FX recovery has too much gain and will clip with a smaller grid signal if the cathode is bypassed.

Here is a suggested circuit:

[Ten Over]

If you were interested in doing it without Merlin's loop, this might interest you:

[xtian]

I’d appreciate any comments/advice/other on the design

IMO, an FX loop in an AB763 build has no benefit or utility.

[pullshocks]

Thanks for starting this thread and for the schematics that have been posted.

I have a 6V6 SE amp project going that currently is pretty close to a Champ AA764--master volume, extra node in power supply and some tweaks to the tone stack and V1A cathode resistor and cap.

I would like to have reverb, but am not sold on spring reverb. I actually prefer the reverb from my Alesis Nanoverb.

A long time ago I had wired in an effects loop but never connected it or tried it out. The design came from the AX 84 web site. It is for the October Studio, a "Marshall Style" amp. I believe Merlin B designed it or was involved with the design.

Yesterday I kludged up the rest of the wiring and tried it out. It sounds fine, but volume and bass response are reduced compared to what I get with the loop bypassed. I suspect it could be improved with some other circuit changes.

Here is an image of the AX84 schematic and a schematic of my amp as of today.

[pdf64]

What benefit do you perceive from the fx loop?
Do you intend to reinstate the NFB loop at some stage? I don’t think that putting the fx loop inside the NFB loop would work out properly.

[pullshocks]

What benefit do you perceive from the fx loop?
Do you intend to reinstate the NFB loop at some stage? I don’t think that putting the fx loop inside the NFB loop would work out properly.

Well, the idea would be to have digital reverb ( I really like the "hall" and "plate" digital reverb sounds, and dislike the way spring reverbs I have tried react to hard picking) applied after tone is shaped in the preamp, i.e. EQ and overdrive. Just as Fender PRs, DRs, TRs, SRs, etc. apply the spring reverb after the preamp. And similar to placing reverb or other time base effects in a loop between preamp and phase inverter of a Marshall style amp. Of course some amps apply spring reverb earlier in the circuit.

I have no actual experience with this. I know for clean playing it works out fine to put the Nanoverb between the guitar and amp. With amp overdrive, I didn't think it sounded that great. Hoping this will be better. Especially after I change the preamp to one with more overdrive.

As far as the negative feedback, that is on the fender schematic but I have never tried it. I have seen other SE amps with FX loops where NFB is applied to the cathode circuit of the loop recovery stage, by adding a 47 ohm resistor between the cathode resistor and ground, and connecting the NFB to the junction of the cathode resistor and added 47 ohm resistor.

[pdf64]

95% of the overdrive will be around the output valve. There’s unlikely to be much overdrive in the preamp, certainly no hard clipping, unless the instrument’s signal is boosted somehow.
So though an fx loop may get rid of the tone control shaping, it shouldn’t have much effect on the overdrive quality, as the output stage would still be mushing all the signal up.

Applying the NFB loop back to the recovery stage means that the feedback will only apply to that signal path, the dry signal path (when the loop is switched out) won’t be affected by the feedback.
So to apply the feedback, you’d need to use an unswitched loop.

[pullshocks]

95% of the overdrive will be around the output valve. There’s unlikely to be much overdrive in the preamp, certainly no hard clipping, unless the instrument’s signal is boosted somehow.
So though an fx loop may get rid of the tone control shaping, it shouldn’t have much effect on the overdrive quality, as the output stage would still be mushing all the signal up.

Applying the NFB loop back to the recovery stage means that the feedback will only apply to that signal path, the dry signal path (when the loop is switched out) won’t be affected by the feedback.
So to apply the feedback, you’d need to use an unswitched loop.

Thank you. Yes I did recognize the issue with bypass and NFB application. If I keep the bypass switch, maybe make it a 3PDT and switch the NFB insertion point at the same time.

This amp has a master volume, and I do not play loud enough to get into power amp distortion.

The Fender style preamp is a recent change. Until recently I had other preamps in there that produced plenty of distortion with the master volume turned way down. I was actually surprised how clean the Fender style preamp stays, even with the gain at 9-10.

So I see your point about loops in fender amps. I see there was another thread a couple months ago
http://ampgarage.com/forum/viewtopic.php?f=8&t=34271 that covered a lot of the same issues.

I'm going to experiment with the loop a little more, but after reading your posts in the thread above, I can foresee moving in the direction of a wet/dry setup. Would prefer to have everything in one compact amp, but...

[pullshocks]

Sorry if I am hijacking this thread, but here is a progress report.

I changed the loop circuit to the one posted by Ten Over in post #3. For the attenuation I used a 500K pot set as close as possible to the 470K/27K voltage divider. New circuit is shown below. Thanks to Ten Over for posting it.

It sounds much better than the other loop I had in there, no loss of bass. I am pretty satisfied with the sound quality.

With Gain on 5, Master Volume on 7, and the loop bypassed, strumming the open strings gave a SPL of about 80 dBZ (an unweighted SPL reading). With the loop unbypassed but nothing plugged in the loop, SPL dropped to about 77dBZ. Maybe adjusting the attenuation would deal with this.

Plugging the Alesis Nanoverb (See attachment for specs), with no attenuation on the 10K send pot, and the Nanoverb input level at noon, the input level indicator stayed in the green, did not flash red. So send level is good.

However, even with the Nanoverb output level at maximum, there is a very noticeable volume loss. SPL was only about 68-70 dBZ, compared to 77 dBZ with nothing in the loop. The specs state the Nanoverb max output level is +17.5 dBU.

I don't know if other effects units have higher max output levels that would avoid this loss of volume. I can get around it by turning up the master volume, but this is not as seamless as I (naively) would have hoped.

[pdf64]

Have you got the Nanoverb set for -20 or +4dB on the front panel?
You’ve got a level pot on the input and the output of the ‘send’ CF buffer - why?
If the one on the output is 500k, that completely counteracts the purpose / benefit of the CF’s low output impedance.

[pullshocks]

Have you got the Nanoverb set for -20 or +4dB on the front panel?
You’ve got a level pot on the input and the output of the ‘send’ CF buffer - why?
If the one on the output is 500k, that completely counteracts the purpose / benefit of the CF’s low output impedance.

Nanoverb does not have a -20/+4 dB switch, front or rear. Possibly you have this mixed up with another unit.

Regarding the pot values, please see the attached. The effects send pot is only 10 K, exactly as shown in Merlin's book. The 500k pot is in lieu of the fixed voltage divider before the triode

[pdf64]

Ahh, that looks better, thanks for the clarification.
The -20 / +4dB switch is noted on the Nanoverb spec sheet that was included as an attachment to your previous posting. Maybe it pertains to an earlier model version than yours?
Whatever, I’d expect the Nanoverb to have unity gain, unless its controls were adjusted otherwise. So I’m somewhat out of ideas

[Ten Over]

With Gain on 5, Master Volume on 7, and the loop bypassed, strumming the open strings gave a SPL of about 80 dBZ (an unweighted SPL reading). With the loop unbypassed but nothing plugged in the loop, SPL dropped to about 77dBZ. Maybe adjusting the attenuation would deal with this.

Plugging the Alesis Nanoverb (See attachment for specs), with no attenuation on the 10K send pot, and the Nanoverb input level at noon, the input level indicator stayed in the green, did not flash red. So send level is good.

However, even with the Nanoverb output level at maximum, there is a very noticeable volume loss. SPL was only about 68-70 dBZ, compared to 77 dBZ with nothing in the loop. The specs state the Nanoverb max output level is +17.5 dBU.

The loop was designed to have nearly unity gain with nothing plugged into the loop when using a 12AX7. You are showing a 12AT7 which will not result in unity gain. You could try bypassing the cathode resistor on the return 12AT7 triode or just use a 12AX7.

The Nanoverb has a maximum output of 5.81Vrms, but that doesn't mean that 5.81Vrms is going to come out of it when the output knob is dimed. It depends on how much the input knob is rotated and what the gain of the Nanoverb is.

The signal level hitting the return triode with nothing in the loop is typically higher than the line level signal coming out of an effects device, so volume loss with a loop is going to be the typical experience.