Help dissecting "One Wire" mod

[Rp3703]

I’m trying to figure out how the gain increases using the “One Wire” mod by shoving a guitar chord into channel 1. All I could think would happen is the 1Meg resistor to ground would get disconnected. Is that it? If you are not familiar with the “One Wire” mod, the instructions, taken from another forum are below. To see it in action look here: https://www.youtube.com/watch?v=CJz6uDdhLuc

This is for Super Lead type amp but you can do it on similar amp also... If you want to maximise this mod do a swap of caps before you begin.....The VOLUME II channel should use the .0022uF for coupling cap and the VOLUME I bright channel should use a .022uF coupling cap...also you should use the tone stack withthe 250pF on the treble with 56K slope resistor...also use .1uF for the PI caps....

Go to the VOLUME II pot....unsolder the wire in the center terminal of the pot wiper...then solder this wire to GROUND, you can just solder it to the next pin on the same pot that is grounded...

the mod itseft:Take a 1 MEG resisitor and solder it to the center wiper.... then take the other end of the 1Meg resistor and solder it to the junction of the 68K input resistors of CHANNEL 1 .........ie you solder it where the two 68K resistors are joined together for channel 1 ... Now yo are done.... now plug guitar into channel II then turn the Channel I and Channel II volume to 10 then keep the MAster Volume to control the volume.... You should now have a good amount of gain and crunch.... Now to add extra gain and crunch you will need a 1/4 male jack only with no cable attached to it.....take this plug and stick it into channel 1 jack and you will now get even more gain.....

For those wanting to try this mod it is imperative to have a Master Volume with your amp, if not it will work but won't sound very good, The MV will limit and control it, etc, Post Phase inverter master volume will do it perfect.....!!!

[HotBluePlates]

I’m trying to figure out how the gain increases using the “One Wire” mod ... the “One Wire” mod, the instructions, taken from another forum are below. ...

Can you provide a link to that other forum? The instructions you reposted (from the other forum) are poor at best, not easy to follow at all.

I haven't listened to the audio with the linked video, but notice the example amp is a JCM800. That amp already cascades the gain stage for the High input into the Low Input channel to raise gain. The mod's information that it is for a "Super Lead type amp" would imply the JCM800 is modded to no longer have the JCM circuit... So who knows what the guy is doing?!

And with all the things being mentioned in the steps, the name "One Wire Mod" seems just a bit misleading...

Separately, input jacks often have a switched contact which opens up when a plug is inserted. If gain is changing with a plug being inserted, then something is being actuated (probably disconnected) by insertion into that jack. But as I said above, the text is poorly written to make clear what the circuit is and what's being changed.

[HotBluePlates]

Something occurred to me watching the video...

I can imagine a guy playing onstage. He gets ready to play a solo, but has to run back to the amp to plug an unused cord into his amp... [/Jerk-Mode Off]

Anyway, overall it looks like the guy took a JCM800, modified it into a non-JCM800 circuit, then did a mod which got him the JCM800 sound again.

[Rp3703]

Sorry for the late reply. I guess I forgot to check the email me replies box. Here is the forum. http://www.marshallforum.com/index.php?threads/for-more-gain-the-one-wire-mod.97/ 

There is supposed to be more info on the Metro Amp forum but it is down right now so I have not seen any of it. There is also this video showing how it’s done https://www.youtube.com/watch?v=W_4pCaG4z8o

I’ve attached a drawing of how I interpreted what was written. If my interpretation is correct then the guitar cable insertion should only disconnect the 68K/1M path to ground. I have a crap amp I’ve been trying out different circuits on that I wired up using this method but with only one input jack, so it would, I guess, sound the way it should if the guitar plug was inserted. The one wire goes directly to grid 2. It does sound really good. 800ish without all the raspy high end fizz. I can’t make it work like as described in the Marshall forum site though. If I crank the Normal volume to 10, it oscillates. I can only go to about 7-8. It also doe not sound very good going lower than 7-8. The High Treble volume is pretty much the gain control. I though about just replacing the Normal volume with a resistor but I still find myself adjusting it even with it’s limited range. I never changed out the tone stack and only have a regular master volume.

[EKDENTON]

Maybe  connecting   your channels together with a short effects pedal patch cord for increased gain would be easier than doing that mod?


http://www.marshallforum.com/index.php?threads/plexi-channel-jumping.6703/

[HotBluePlates]

Got it. Thanks for taking the time to draw some notes on the schematic, as the instructions are really for a non-tech soldering parts into an amp while I think in terms of a schematic which shows what the circuit is doing.

... If my interpretation is correct then the guitar cable insertion should only disconnect the 68K/1M path to ground. ...

You actually sorted the answer out for yourself!

Signal path is like this:
1st gain stage with 820Ω cathode resistor & 250µF cap, with 0.0022µF coupling cap
1MΩ Volume control -> Wiper to jacks & grid of 2nd gain stage.
2nd gain stage with 2.7kΩ cathode resistor & 0.68µF cap, with 0.022µF coupling cap
1MΩ volume control with 5000pF bright cap -> wiper to another voltage divider of 470kΩ/470kΩ resistors (with a 500pF bright cap)
3rd gain stage
Cathode follower tone circuit driver


The volume controls are voltage dividers. The jacks on the output of the 1st volume control are an "extra load" which changes how the 1MΩ pot divides voltage.

With nothing plugged in those jacks, the switches on the jacks place the 68kΩ resistors in parallel and connected to ground. The net result looks like 34kΩ to ground.

Let's say you set that first volume control to divide the voltage by half, with 500kΩ from input to wiper and 500kΩ from wiper to ground. The jacks and their resistors are in parallel with the lower half of the pot from wiper to ground, so that 34kΩ is in parallel with the pot's 500kΩ. The total looks like ~32kΩ to ground. Instead of reducing the voltage to 50% of the input voltage, this volume pot setting reduces the voltage to 32kΩ/(500kΩ+32kΩ) = ~6% of the input voltage.

Now you insert a plug into the correct jack with the unused cord, it opens the jack's switch which un-shorts the 1MΩ resistor; the two 68kΩ resistors are still in parallel. The total resistance rises to 1.034MΩ instead of the previous 34kΩ. In parallel with 500kΩ from the pot's wiper to ground, this looks like 337kΩ to ground. With the same volume pot setting as before, output climbs to 337kΩ/(337kΩ+500kΩ) = 40% of the input voltage.

[HotBluePlates]

... If I crank the Normal volume to 10, it oscillates. I can only go to about 7-8. ...

That's not your fault, it's the mod.

Overall, this is a kind of half-baked idea for modifying an amp while only changing the order of parts rather than changing the parts or circuit design itself. Which makes sense if we assume the creator of this change, and the intended audience, have access to a printed circuit board amp which then locks in certain circuit blocks. The voicing of channels is kind of silly (swapping in the small coupling cap to offset the mud of the Normal channel & vice versa) and the voltage dividers on top of voltage dividers, as well as bright caps on top of bright caps (in the 2nd volume control & voltage divider)... It all screams a new amp tinkerer not yet ready to replace the circuit board wholesale with something new or to really design a circuit.

But I'm being too curmudgeonly... Everyone starts somewhere, and this is a good start for a beginner taking control of their amp's sound. It will oscillate, because higher-level signals are being run back to the input jacks next to low level signals, and I bet the circuit crosses over itself in other places from the board to the volume controls.

If this was the end-all be-all of tone except for oscillation, we could redesign the circuit to make more sense & use fewer parts. But since you report it only sounds good over a narrow volume range, and since it is basically 3 gain stages plus a cathode follower (plus phase inverter), I'd recommend looking around for good-sounding circuits with 3-4 gain stages to copy. There's quite a few of them out there...

[EKDENTON]

I’m trying to figure out how the gain increases

If all your really looking to do is increase the gain, try jumpering the channels and see what you think.....just saying.


plug into Bright 1, Jumper from Bright 2 to Normal 1.

[HotBluePlates]

True, but jumping channels is parallel operation, or Gain 1 + Gain 2.

The discussed mod is cascaded operation, or Gain 1 * Gain 2.

But it also throws away so much gain in various voltage dividers, it makes me question how well it works.

[Rp3703]

Sorry, again, I'm only getting notices of replies sporadically for some reason. This mod is supposed to be one of the more common mods done to plexi's back in the day and eventually led to Marshall designing the JMP. I'm not positive but it may also be called the Jose mod. I agree that it is a pretty sloppy cascaded gain structure but I think the purpose was to give you the most with the least amount of work. As for the jumping inputs go, that will not cascade the first two tube stages. It only blends them. Same amount of gain but different tones. That is true, the way the input jacks are wired, you would be running through two parallel 68K resistors and then though the 1M to ground but by plugging in the guitar cable, you only lose the one 68K resistor. I did not see that. So I guess I need to add that to hear exactly how it should sound. I also think I put my gain stages in reverse order. I have the 2K7/.68 first and the 820/250u second. I need to reverse those as well. I did say that the volume one has a limited range of usability but volume 2 has no limits. It can go from 1-10 and pretty much is the gain knob. My bigger problem right now is tracking down the cause of my 120hz hum but that will be for another post.

[Rp3703]

I apologize again for my bad reading of the schematic. I guess there is both a straight to ground and a path through the 1M resistor to ground after the parallel 68K resistors but inserting a guitar plug, I would think, would actually break the connection of one of the 68K resistors since the other end of the cord is not connected tip to sleeve. This would make the resistance go from 34K ohm to 68K ohm. I know there are plenty of other cascade preamp designs out there but I'm kind of liking the sound of this one. Although I have it wired wrong at the moment, I think it is a lot better than a Plexi, even with an overdrive pedal and less fizzy and thin sounding than an 800. I'd be willing to try any suggestions you have on how to eliminate any of the sloppiness though.

[HotBluePlates]

I apologize again for my bad reading of the schematic. I guess there is both a straight to ground and a path through the 1M resistor to ground after the parallel 68K resistors but inserting a guitar plug, I would think, would actually break the connection of one of the 68K resistors since the other end of the cord is not connected tip to sleeve. ...

See the attachments.

There are switched contacts on all jacks, but one of them will bypass the 1MΩ, so only the parallel 68kΩ resistors are seen by the circuit ("No Plug"; 34kΩ total resistance).

When a plug is added to the correct jack, the contact which shorts the 1MΩ is opened (yellow "X" for this opened contact). The previous 34kΩ for the parallel 68kΩ resistors is now in series with the 1MΩ, so 1.034MΩ total resistance. ("With Plug")

[Rp3703]

I get what you're saying but the actual wiring of the inputs is more like the attached drawing which has the sleeve connected to ground no matter which one you choose.

[Rp3703]

Ok so this threw me for a loop but I was doing some searching last night and came across another web site https://sites.google.com/site/yourtubeamp/mods-and-maintenance/-marshall-1-wire-mod
that says this is the actual “One Wire” mod method. Now after seeing the videos and the forum dealing with the other method, I have trouble believing this one but the new signal path is much closer to what became the JMP/800 design. I really wish Metro Amp forum would get their site up and running again.

[HotBluePlates]

I get what you're saying but the actual wiring of the inputs is more like the attached drawing which has the sleeve connected to ground no matter which one you choose.

Yes, the sleeve is the heavy black lines on the schematic, and is always connected to ground.

The switched contacts are the arrows on the schematic. For each pair, one switched contact is connected to ground, while the other is connected to the non-grounded end of a 1MΩ resistor. NOTE: Each Cliff-style jack has a switched contact on the sleeve connection, but it's not being used for anything in this amp.

The remaining bit of the jack schematic symbol is the tip itself.

[Rp3703]

Ok, so to imitate the guitar plug gain increase, I could just run a 1.034M resistor to ground from the grid of the second tube stage or install a switch that offers 34K or 1.034M to ground connections.

[HotBluePlates]

Yes, you could do it that way.

Or you could figure the gain loss through the volume pot/voltage divider circuit in both settings and add a resistor in series with the input of the pot. That resistor, when the volume pot is at maximum, would define an "always-on" level of voltage division/signal reduction. A switch to add a parallel resistor could then reduce the effective value of that series resistor, which lessens the voltage division, and "boosts" signal level. It would be cleaner, with fewer parts.

Re-arranging the layout, or the physical sequence of the gain stages, volume control, etc will be the most effective way to combat the oscillation you mentioned previously.

Is the one-wire amp mod the best Marshall-style distortion you've heard? Is the goal of this thread to duplicate it, or design something which "does the same thing"?

This is my recurring question because I'm still not convinced it's the best cascaded gain-stage Marshall plan out there. And if it's flawed from the outset, why work to perfect it?

If instead you have a different Marshall plan that's "too gainy" or "too raspy" those are easy things to fix. Mostly because super high-gain amps first lop off bass because at stage volume it sounds muddy and ill-defined. Then they tend to shave treble because with high distortion the sound can get harsh/buzzy and hissy. And eventually, you get left with the Mesa sound of mids-only.

[Rp3703]

I'm not ready to say this is the best Marshall distortion I’ve ever heard since I’ve heard some good ones on youtube, but I like the Plexi tone over the 800 but even with an overdrive pedal in front of it, it is still lacking to me. I guess I’m looking for some middle ground between the two. 
I have this crappy home brew 100W tube amp I built that is set up to where it’s pretty easy for me to change the pre layout with a limited amount of work. Out of curiosity, I have been using it to try out different Plexi mods I’ve found online. I tried a #36 I found but that ended up sounding like crap. This one sounds pretty good so far but I just rewired it the way, I think it was supposed to be wired and it sounds way less gainy than the screwed up way I had it wired before. Not sure if that is due to swapping the order of the gain stages or from adding the 34K/1.034M voltage divider. I guess I’ll remove the divider first and see what that does.

[Rp3703]

So I just removed the voltage divider and it has even less gain. So this is leading me to believe that putting the 2K7/.68 stage first is the best arrangement of gain stages.

[HotBluePlates]

... putting the 2K7/.68 stage first is the best arrangement of gain stages.

Perhaps.

If you do that, the 250µF bypass cap of the 2nd stage will be silly to have. That's because the 1st stage will have stripped out much bass, so there will be no point in going above 10-25µF in the 2nd stage.

[Rp3703]

Sorry to revived a 4 year old thread but since my thread shows up fourth in line when I google "One Wire Mod", I felt it necessary to address all the inaccuracies I wrote so anyone else who may come across it looking for some understanding does not get the wrong info.

First off, I have no idea why my original post included a Youtube clip of a JCM 800. This mod is meant for pre-2203 marshall circuits(it can be performed on many other amps but that's a whole other subject) with normal and bright channel inputs. It takes the separate bright and normal channels and runs one into the other in a cascade arraignment. The JCM 800/2203 is already cascaded.

Second, the OWM is not the same as "jumping", which involves running both channels at the same time in parallel.

Third, I do not believe I actually heard how the OWM circuit sounds. I'm not exactly sure what circuit I had wired up when I made my comment about this being "the best Marshall sound I have ever heard" but I do know It was not true to the pre 2203 circuit design.
 
Fourth, to this day, there is still conflicting information on how actually the OWM is supposed to be wired. As I posted back in 2016, this forum http://www.marshallforum.com/threads/for-more-gain-the-one-wire-mod.97/ states that the signal runs through the normal gain stage first and then cascades through the bright channel gain stage. This Youtube video https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DpCESJJ0gGro&psig=AOvVaw3hZEbrhThc-615hnBjNZtX&ust=1591706971185000&source=images&cd=vfe&ved=0CAMQjB1qFwoTCPD1mdeg8ukCFQAAAAAdAAAAABAD  shows how it's done.

Then if you look on this website, http://www.lynx.net/~jc/marshallAmps.html, it shows the signal going through the bright channel first and then through the normal channel. This route is very similar to how the JCM 800/2203 is wired. It would make sense that Marshall just copied what amp techs were already doing to their amps. In fact if you look at the first productions of the 2203, they are built onto old 1959/87 circuit boards.

Since the mod is called "One Wire" it makes sense to me that all there is to the mod is one wire running from junction of the two 68K grid stop resistors on either channel to the middle lug of the other channels volume pot and that same channels existing center lug wire being run to ground. Other than that, my thread should not be used as a source of info on the subject. Sorry for the misinformation.