Seymour Duncan Convertible 100 Input

[bobloblaws]

The input on this Seymour Duncan Convertible amp (schematic and manual here:Seymour Duncan Schematics - Tube amp Schematics had faulty jacks and I have replacements but I'm confused as to the correct way to wire it back up. The schematic shows a 4.7K input/grid leak resistor. But the manual states that "Convertible amps have 4.7 meg-ohm input jacks". Which one is the typo? (Strangely, this particular amp had a 170K input resistor installed instead.) The manual then goes on to say "To optimize signal to noise ratio, input impedance should be fairly low - typically 100 to 10k ohms. Try the lOk330pf load resistor plug for great tone and low noise." Huh? Isn't the input impedance roughly equivalent to the resistance on the input jack (4.7M or 4.7K depending on which document you prefer)? The amp has an RCA jack for the aforementioned "load resistor plug" which is wired between the main input jack and the first input module. The RCA jack is not shown on the schematic, nor is the second input jack. I'm pretty sure this is the correct schematic, everything else lines up and I haven't seen any different revisions lurking around the net. I can' t tell if the two input jacks are supposed to be in a typical Hi/Lo configuration. There also doesn't appear to be any grid stopper resistors, what's up with that? Anybody have any idea how this should be wired up?

[bobloblaws]

So no replies on this as of yet and I don't blame you guys. For one thing I was barking up the wrong tree, at least in part. And this is one mighty unconventional amp. Anyway, I've made some progress in determining what is up with the input section but I'd love to know why it is the way it is. To cut to the chase, there is no typo in the manual or schematic. The input impedance is in fact 4.7 Meg as stated in the manual. I discovered this by measuring the input resistance before attaching a grid leak resistor. The meter read ~ 4.9 Meg. My understanding was that the grid leak/input resistor determined the input impedance. But obviously Seymour Duncan accomplished this another way. Can someone enlighten me? I removed the 5 pre-amp modules and still got the same reading, FWIW.

[Leo_Gnardo]

The input impedance is in fact 4.7 Meg as stated in the manual.

4.7 Meg makes sense. One of the unique features of the SD Convertible, is an RCA jack in parallel with the regular input jack. This is not intended as an alternate input. Instead it's a receptacle for RCA plugs loaded with loading resistors. I dunno, maybe Duncan offered ready made ones. Or it's easy enough to make your own with cheap RCA plug shells. Seymour and/or his amp designer were ahead of the rest of the crowd, recognizing that the load resistance a guitar output "sees" at the amp input has an effect on tone, and AFAIK they were the only outfit to offer a way to vary the input load resistance of an amp this way.

Along the same line, there's at least one commercially available "brown" box offering variable loading, and in 1993 I made one of my own. A unique "effect" in that it requires no battery, and either jack serves as input or output. The lower you dial the resistance, the more high & low frequencies are cut from the signal by way of impedance mismatch. "Dial down for brown!"

[bobloblaws]

4.7 Meg makes sense. One of the unique features of the SD Convertible, is an RCA jack in parallel with the regular input jack. This is not intended as an alternate input. Instead it's a receptacle for RCA plugs loaded with loading resistors. I dunno, maybe Duncan offered ready made ones. Or it's easy enough to make your own with cheap RCA plug shells. Seymour and/or his amp designer were ahead of the rest of the crowd, recognizing that the load resistance a guitar output "sees" at the amp input has an effect on tone, and AFAIK they were the only outfit to offer a way to vary the input load resistance of an amp this way.

Along the same line, there's at least one commercially available "brown" box offering variable loading, and in 1993 I made one of my own. A unique "effect" in that it requires no battery, and either jack serves as input or output. The lower you dial the resistance, the more high & low frequencies are cut from the signal by way of impedance mismatch. "Dial down for brown!"

Yes, according to the manual the amp came with 2 pre-made load resistors. Am I correct in my understanding that typically the input impedance is set by a high value grid leak resistor, as in vintage Fender/Marshall amps using a 1 Meg grid leak/input resistor? And how is it that the amp in question has an input impedance almost 5 times that of said Fenders and Marshalls despite no input resistor (aside from the puny 4.7K resistor shown on the schematic. Incidentally, the amp came to me with a 170 K input resistor, I'm not sure why someone picked that value when so far my experimenting shows that a much lower value results in much less hum, I tried the 4.7K and then got even less hum by going to 2.2k).

[Enzo]

I think you are overlooking the input stage module. You can select from a variety of them, but they are in parallel to that resistor. A number of them have like a series 100 ohm and a 1 meg to ground at a grid. SOme have a cap in series, mainly the solid state ones.

So they may well have "4.7M input jacks", but that is not the input impedance the guitar sees.

[g1]

I got 2 different schematics from that link you posted. One shows a 4.7K, I think that's probably a typo.
The other shows a 4.7M, and ahead of it, something labelled "RL" which I take to be the loading resistor as Leo called it, which would connect via the RCA jack.

[Leo_Gnardo]

Am I correct in my understanding that typically the input impedance is set by a high value grid leak resistor, as in vintage Fender/Marshall amps using a 1 Meg grid leak/input resistor?

Yes, the grid leak resistor and input load resistor are one and the same. Hi sensitivity input, #1 in the usual 60-70's Fenders, 1 Meg is the value. #2 input puts the 68K input mix resistors in series as a load, and the 1M is in parallel with both, for about 125K. 60's vintage Ampegs often used 5.6M resistors as load/gridleak. Also keep in mind, once an ordinary - no preamp - guitar is plugged in, then its own output resistance swamps the 1M (or whatever) and becomes the grid-leak itself; dial the volume to zero and then the only gridleak value left is that of the stopper resistor. If there's a cap blocking DC connection between the first preamp tube's grid and input jack, then the value of the existing gridleak R remains constant, but this is a seldom seen circuit, usually in old amps with grounded cathodes in the input triode circuit.

[GrahamPearson]

I own 2 100 watt Convertibles. One was manufactured in 1982 and the other in 1987. The later one has had all of the modifications applied suggested by the various service bulletins.

First, I can confirm that both amplifiers have a 4.7MΩ resistor in the input section. Yes, there's an error on most schematics. Unfortunately, there are other errors, both on the PCB and the schematics. Here's what I've found.

On the PCB, there are two resistors labeled R58. The first references a 470K 1/2W resistor; the second references a 220K 1W resistor. Capacitor C30 is labeled “0”.

On the various versions of the schematic:
Variable power tube V3 is labeled 12AX7 instead of 12AU7.
Resistor R7 is labeled 2 W instead of 5 W.
Resistor R10 is labeled 4.7K instead of 4.7M.
Resistors R51 (100Ω) and R52 (100Ω) are omitted from the connection between pin 8 of V3 and pin 3 of V3.
Resistor R61 is labeled R60.
Resistor R64 is 470Ω 1/2W, not 470K 1/2W.
Unlabeled resistor is labeled 100K instead of 10K on sheet 2 of 2 (top-right corner).
Capacitor C28 is 0.1µF/400, not 1µF/400.

I hope this helps.
Graham

[bobloblaws]

I think you are overlooking the input stage module. You can select from a variety of them, but they are in parallel to that resistor. A number of them have like a series 100 ohm and a 1 meg to ground at a grid. SOme have a cap in series, mainly the solid state ones.

In fact, I removed all 5 input modules and still got the same measurement.

So they may well have "4.7M input jacks", but that is not the input impedance the guitar sees.

I don't follow this. Why would it make a difference in what the guitar "sees" if the input impedance comes from a resistor tied physically to the jack or from components mounted in one of the modules? And if my ohmmeter can see it why wouldn't the guitar be able to see it?

[bobloblaws]

I own 2 100 watt Convertibles. One was manufactured in 1982 and the other in 1987. The later one has had all of the modifications applied suggested by the various service bulletins.

First, I can confirm that both amplifiers have a 4.7MΩ resistor in the input section. Yes, there's an error on most schematics. Unfortunately, there are other errors, both on the PCB and the schematics. Here's what I've found.

On the PCB, there are two resistors labeled R58. The first references a 470K 1/2W resistor; the second references a 220K 1W resistor. Capacitor C30 is labeled “0”.

On the various versions of the schematic:
Variable power tube V3 is labeled 12AX7 instead of 12AU7.
Resistor R7 is labeled 2 W instead of 5 W.
Resistor R10 is labeled 4.7K instead of 4.7M.
Resistors R51 (100Ω) and R52 (100Ω) are omitted from the connection between pin 8 of V3 and pin 3 of V3.
Resistor R61 is labeled R60.
Resistor R64 is 470Ω 1/2W, not 470K 1/2W.
Unlabeled resistor is labeled 100K instead of 10K on sheet 2 of 2 (top-right corner).
Capacitor C28 is 0.1µF/400, not 1µF/400.

I hope this helps.
Graham

Thank you Graham, that is very helpful indeed. But I'm still a tad skeptical on the input resistor. Is it possible that previous owners or techs came to the same conclusion that the "official" schematic has a typo and installed a 4.7M resistor based on the spec listed in the manual? (I'm assuming the schematic with the 4.7K resistor is "more official" than the more roughly drawn one. And in the file seymourduncan_convertible_100.pdf the 3rd page shows a date of 1994 vs. 1984 in the file Seymour_Duncan_100w_Conv.pdf)
The fact that I get a ~4.9M reading on the input without a resistor installed (or the 5 input modules) reinforces my skepticism. Do the input resistors in your amp look to be original factory installed?

[Enzo]

If you pull all the modules, then all you will measure is the 4.7 meg resistor. You are measuring resistance. The guitar signal sees impedance, not resistance. Depending on the module installed first, you might see a 100 ohm plus 1 meg parallel your 4.7 meg. That affects the impedance the guitar sees. Other modules have an input cap. That blocks your meter from measuring anything ther, but the cap is coupling the guitar signal into the module, and the impedance there will be part of what the guitar sees. In other words, that 4.7 meg resistor is not all the guitar sees.

An example is in your guitar, the tone control. Your tone control has a cap in series with it, so if you measure resistance at the guitar jack, turning the tone knob won't vary the reading, but the signal still passes through that cap, and so the control changes still affect the signal.

Typos can appear in any schematic, not just older ones.

[bobloblaws]

If you pull all the modules, then all you will measure is the 4.7 meg resistor.

But there is no resistor. I measured before hooking up an input resistor. That is the whole point of the question I am asking in post #2.

[Enzo]

Your meter is measuring something, such a resistor could be inside the jack, on the rear of the main board, on the bottom of the module 1 edge connector, etc. Have you had the main board out?

[bobloblaws]

Your meter is measuring something, such a resistor could be inside the jack, on the rear of the main board, on the bottom of the module 1 edge connector, etc. Have you had the main board out?

Ahhh, good call Enzo, I see it now on top of the board beside the two posts where the input comes into the PCB. Duh. It didn't occur to me to look on the board for some unknown reason, my "thinking" process was it's either on the jack itself as in a vintage Fender amp or in the first input module and if not then it's a mystery for the ages :-)

Edit: Actually, I know what threw me off. Because there was some question of which version of the schematic had a typo, I speculated that the 4.7K resistor was correct (as I said there was a 170K installed and I found that changing that to a 4.7K reduced a lot of hum), so then I thought if my meter shows 4.7M and it is not shown on the schematic it must be in one of the "black box" input modules. But when I removed them and still got the same reading I was scratching my head. Anyway, it all makes sense now, thanks everybody!

[bobloblaws]

On the various versions of the schematic:
Variable power tube V3 is labeled 12AX7 instead of 12AU7.

Hi Graham, please disregard my comment about being skeptical of your assertion regarding the 4.7M resistor, I am no longer skeptical, I've now seen it with my own eyes :-)

But I'm wondering about V3. How can we know for sure V3 is supposed to be 12AU7 instead of 12AX7? Incidentally, with the amp in question, it looks like the owner or someone else might have been wondering about this too as it actually had a 12AT7 installed in that position, splitting the difference, as it were.