FX loop MIX control help

[SILVERGUN]

I have this amp wired up on my bench and this is the first time I am trying this specific idea of a parallel MIX control on an effects loop.
It would appear as though my FX recovery signal should be out of phase with my dry signal but I'm not hearing a problem.
The MIX control is based off of an example in Merlin's 1st edition preamp book, so I don't think he would have suggested it if there would be a phase issue.
I don't currently have my scope hooked up to look at it.
Can you guys take a look at it and let me know if you see any problems with the implementation?

[acheld]

OK, I'm not the pro here, but it looks good to me.  There "shouldn't" be a phase issue unless the effect introduces one.

Looks like a cool project.

[SILVERGUN]

I thought of my own experiment so I went down and took the delay/reverb effect out of the loop so that I would be able to hear the effect of being out of phase.
At one end of the mix control I get the full dry signal coming from the relay. At the other end of the pot sweep I get the full signal from the FX return, and right in the middle of the sweep I get full phase cancellation.
With the delay/reverb pedal back in the loop I don't get full phase cancellation at the midway point in the pot but I do get a definitive volume drop. So I see now why I wasn't seeing a problem. (because I wasn't moving the pot while playing)
The good news is that it does work well with the effect full up and the MIX control leaning way more towards the dry signal I get a beutiful, semi-effected sound which is very nice.
Any issues with using it like that?
Any suggestions for improving the design?

[sluckey]

It would appear as though my FX recovery signal should be out of phase with my dry signal but I'm not hearing a problem.

That's true if there is nothing plugged into the send/return jacks. The mix control should tend to cancel the signals in that case. But when you plug something into the jacks all bets are off.

[SILVERGUN]

Well that's good news. It definitely sounds better with the pot leaning way more towards the dry signal with the big effect in the background.
This will be my amp and I'll be running some kind of reverb all the time.
What happens when you mix some out of phase signal intentionally?...just volume reduction?
Is that what goes on in a crossline master volume?

[tubenit]

tube effects loop on board (el34world.com)


There are probably a dozen plus examples of FX loops in ARCHIVES thread.

Posts #24 and #27 show examples of using a potentiometer to switch from series to parallel loop.  Don't know if you're already familiar with that or not?

To my ears, I do hear a distinct difference between a series loop and a parallel loop.


With respect, Tubenit

[shooter]

when you plug something into the jacks all bets are off.

unless you scope each effect, you cannot know phase of both amp and effect.

[pdf64]

It would be beneficial to try to avoid conflating signal phase with signal polarity.
Unless we’re only dealing with single frequency signals, inverting signal polarity is a fundamentally different thing to a phase shift.
Polarity can be flipped without phase being affected.
Whereas phase is a time / frequency dependent characteristic.
The fx loop under discussion flips signal polarity. Signal phase, within the frequency range of interest, should be largely unaffected. 
The point being that as a delay fx is being brought into play, phase really will be getting messed with.

I think it would be unusual for the dry signal path of modern delay etc effects to have their polarity flipped. 

The operation of a type 3 (crossline) master vol seems non trivial. Different things probably come into play in the different ranges of control setting.
But signal cancellation per se doesn’t occur. The amp mutes because the LTP outputs become common mode, rather than differential.

[shooter]

 

conflating signal phase with signal polarity.

they've been conflated since I joined this group, just using the majority phrasing

[SILVERGUN]

Thanks for the help guys. Instead of quoting everyone I think I can make this pretty simple.
shooter - yup, it would be helpful if I was able to see this on a scope as we're talking about it. I should be able to get hooked back up this week.
tubenit - as always, thank you so much for pointing me in the right direction. That thread you linked is full of great info. and I've got some reading to do.
pdf64 - I always appreciate keeping the terminology straight and I see what you are saying. And thanks for addressing my question about the crossline master. I need to learn more about phase so that I'll be less likely to use the term incorrectly.

I've had the luxury of a learn-as-you-go experience here and it's time for me to learn more about this specific subject. There is a hole in my knowledge and I know I've got some reading to do. I have heard the effects of phase issues but I'm not informed enough to understand whats going on.

From a knob twisting, uninformed guitarists point of view I'm glad I installed it like this. The "morphing" effect of the signal getting close to and then flipping polarity that I (think I) am hearing throughout the pot sweep was enough to get me interested in the answer.
On the far end I have what it would sound like with the loop in series and the effect all the way up, which is terrible.
On the other end I have the big, bold, uneffected dry signal. In the middle they somewhat cancel each other out. Turning the pot from all dry slightly towards mildly effected gives me the best result.
I have the big, dry sound with a perfectly mixable reverb/delay giving me the depth I wanted but without washing out the dry tone.
This would be a bad idea on a commercial amp because some dope would put the pot in the middle and then tell the world how much the loop sucks.

I'll continue to experiment with it and push to get my scope hooked back up so I can see what I'm hearing. I'll also read some more and see where that takes me.

[pdf64]

Are you sure that the mix circuit Merlin suggested operated the same as your’s, ie such that dry signals of opposing polarity were at either end of the mix control?

[SILVERGUN]

Are you sure that the mix circuit Merlin suggested operated the same as your’s, ie such that dry signals of opposing polarity were at either end of the mix control?

Here's where I got the general idea from. I didn't have room for separate pots, but the polarity issue remains the same either way.


**Let me know if you think Merlin would not want this posted and I'll take it down. Mine is more "inspired by".

[SILVERGUN]

My main concern is:
Ideally, wouldn't we want an additional stage after the recovery triode, but within the loop, to flip the polarity to match the dry signal at the MIX control?
And if not necessary, what am I missing that is happening in the effects unit that makes my concern unwarranted?

And I apologize. More knowledge on my end would make this less painful.
Thanks!

[shooter]

under NO FX condition, your ears and signal don't care, there is nothing to be "flipped"  IF!! you don't jump send to return.


in FX signal mix condition, there is no way to know what the external "things" do to signal polarity, or for that matter phase, so you get what you get.


I experimented with actual phase shifting within the AMP.  Measured each stage, compared phasing to original signal.  9/10 times the amps/circuits would shift phase between 0 and ~~~~25 degrees.  homebrew, store bought, there were always some shifting.

then HBP chimed in, and PRR, and Sluckey.  They all stated talking about the sound coming from the speaker, the shifting and mixing and complexities of sound bouncing and, well........
I got properly skool'd on phase in guitar world

[SILVERGUN]

I got properly skool'd on phase in guitar world

It's time for me to go back and read Basic Audio 1,2 and 3.
This is what happens when a welding machine R+R guy tries to get fancy with audio. 

[shooter]

I came out of the electronics world, was well trained.
in that world signal phase was EVERYTHING.  so it took me some time and head scratching to accept that in this world, it's a sub-minor irritant at best, ignored as a general rule.


if you start designing audio for the cork folk, then maybe spend some time on studying ELI the ICE man. how caps and inductors shift phase and ways to mitigate that shift.  But if you go down that rabbit hole, expect the next place you pop out will be sound spectral analysis.  THAT hole will require some better herbal remedies than even I can get 

[SILVERGUN]

I'm thinking about trying this next.
Any objections?
Comments?

[66Strat]

I'm thinking about trying this next.
Any objections?
Comments?

I like it. I don't see a problem.

[pdf64]

It looks fine but you lose the global nfb on the power amp?

To avoid the nulling behaviour of the mix control, the original circuit could be changed so that the send buffer was an inverting type, eg

[66Strat]

It looks fine but you lose the global nfb on the power amp.

The original circuit could be changed so that the send buffer was an inverting type, eg

He would have to add a global NFB loop before he could lose it.

[SILVERGUN]

It looks fine but you lose the global nfb on the power amp.

The original circuit could be changed so that the send buffer was an inverting type, eg

He would have to add a global NFB loop before he could lose it.

It was there on the first schematic. That will probably change at least 6 more times before I'm done. 

[SILVERGUN]

It looks fine but you lose the global nfb on the power amp?

To avoid the nulling behaviour of the mix control, the original circuit could be changed so that the send buffer was an inverting type, eg

I was considering that. I don't need any more gain and I have to attenuate the signal pretty hard as-is.
Maybe make the loop 12AU7 and use a plate fed send..(?)
I don't plan on running long cables in the loop, so the impedance benefit from the cathode feed is not totally necessary.

[pdf64]

Even with a 12AX7, the circuit in post #19 will have an output impedance less than 1k.

[shooter]

It might be nice to show your work getting to 1k for those that are scratching their heads and trying to google "output impedance"

[pdf64]

I was going to add this link in the next ad break 
http://aikenamps.com/index.php/designing-single-stage-inverting-feedback-amplifiers

[SILVERGUN]

It might be nice to show your work getting to 1k for those that are scratching their heads and trying to google "output impedance"

I didn't even get there.
I did however get to the point shown in reply#16 and just when I was starting to have fun I got shot down by the man.
I was trying to tune in the PI input coupling caps with a substitue box and I got the kibash from upstairs.
I am tweaking heavily and wound up adding a mid-control on the OD channel with just a trim pot on the board.
The combination of that plus the new PI input arrangement has me smiling.
I need some time to get the schematic up to speed.
Are there any special considerations to balancing the PI now that the NFB is gone?....since I'm using a 12AY7?....and since I'm feeding both inputs?
Has anyone ever used grid stoppers on the PI grids?  How would that affect PI setup/operation?

[Willabe]

I'm thinking about trying this next.
Any objections?
Comments?

That's a cleaver work around. (Reply #16)

If you want/need FB, use a UL OT instead of a -FB loop. Still use 1K screen grid R's with the UL screen taps. UL OT's get a bad wrap, guys think the UL OT was what made later SF Fender amp sound thin/bad, but it wasn't the UL OT's, Fender added those power tube grid to ground bleeder caps and buggered some of the circuits. The UL screen connection taps not only help protect the screens but also provide a fair amount of FB.

Besides, the Fender UL OT's weren't true UL, they were only something like 14% or 18% taps. True UL OT's have something like ~40% taps?

[tubenit]

When you get this all done, I would love to hear a soundclip please!   I think your FX approach is interesting and I like the idea alot. If you can post a soundclip, please do.


With respect, Tubenit

[SILVERGUN]

Thanks Willabe, my decision to get rid of the NFB was made easier by this comment from Aiken:
Negative feedback makes the amp sound "tighter", particularly in the low end, where the speaker resonant hump has the most effect on amplifier output. This is better suited for pristine clean playing or a tight distorted tone, while a non-negative feedback amp has a "looser" feel, better suited to a bluesy, dynamic style of playing.

Since I'm more sloppy than pristine and I am going for a looser, bluesy, dynamic style  , this was a no-brainer.
Also, while I was tweaking the original NFB pot I found that I liked the result of less and less NFB. It was just a little tough to get used to the added noise that was being induced. Now that it's gone I am enjoying the amp more and I'm getting used to the added noise.


Tuning in the PI input cap value was a huge ah-ha moment for me last night and I'm looking forward to getting back to it tonight.
I might add the grid stoppers as insurance while I'm in there. I have a feeling that once I button this up I'm not gonna stop playing through it long enough to go back in. I'll start another thread with some build pics and more info about the tweaks, etc.

[SILVERGUN]

When you get this all done, I would love to hear a soundclip please!   I think your FX approach is interesting and I like the idea alot. If you can post a soundclip, please do.


With respect, Tubenit

Absolutely.
This is the best I have done so far. It's getting closer and closer everyday now and I'm proud of how I have been able to shape it into what I wanted it to be.
Thank you again and again for all of your direction in the past. Your guidance has given me the freedom and confidence to do what I'm doing and truly enjoy this hobby.
The loop work is mostly from Merlin's Preamp book and it was his idea to use the other half of the PI to inject the FX return signal.
There was just too much cancellation going on the MIX control the way I had it originally.
Now, I have one volume control of dry signal and one volume control of effected signal and I didn't realize that's what I would want.
But boy does it ever wok great!

[66Strat]

I look forward to hearing it as well.

https://www.youtube.com/watch?v=FPQlXNH36mI

[Willabe]

Thanks Willabe, my decision to get rid of the NFB was made easier by this comment from Aiken: ....

Ok, carry on.

Just remember that not only do KT88, 200w, Marshall Majors have UL OT's, but they have a 12AU7 after the PI driving the KT88's, and you like that. Our friend Ritchie200, Jim, would agree, gotta have it! 

For anyone who might want to explore UL for feed back instead of a -FB loop from the OT's secondary, from PRR in another thread;

"But this is a guitar amp? Mostly, heavy damping of a guitar stage amp makes it "lifeless". The classic Fenders ran about 6dB NFB. Using UL tends to give roughly 6dB NFB, boom, done. (Except you don't have a place for crazy NFB tricks like Fender's Presence knob.)

UL *tends* to survive stage abuse a bit better than Pentode (yet louder than triode). When the plate pulls down, G2 is pulled down too, which limits the peak current. IMHO the Fender "135W" on four bottles would have to be rated 100W if not for the UL connection."

[SILVERGUN]

Here's a link to a sound clip demonstrating the dry vs effect signal in this amp.


https://soundclick.com/r/s8karx

[SILVERGUN]

Here's an updated schematic. I'm still working on the OD channel but haven't had a chance to get it to a decent volume. I think the clean channel sounds really good.

[shooter]

haven't had a chance to get it to a decent volume.

you always have a chance, it's the consequences you're worried about 

[66Strat]

Sounds really good.

[SILVERGUN]

Thanks JT!
I just got done with some more tweaking and I'm getting close. I fell in love with the idea of the AY7 in the PI, but to me it sounds much better with an AU7 in there.
The OD needed to be tamed since I dropped global NFB and I just added a .001 cap right before thd PI entrance cap to ground on the dry signal side and that really helped.
I tried a bunch of different locations and that just worked out as the winner.

haven't had a chance to get it to a decent volume.

you always have a chance, it's the consequences you're worried about 

The good news is, she's starting to hum along.

[SILVERGUN]

For anyone who was playing along and might be interested I have this to share.
My new delay/reverb pedal came in and it is compatible with an app where you can change a bunch of settings/store presets, etc.
One of the settings that you can change says:
"invert output signal" 
I was not expecting to see that.

[shooter]

or just turn the guitar upside down