The "capacitor trick" on a Superblock US: 3 values compared (with audio)

[nevershouldhavesoldit]

[This discussion and method are applicable to any amplifier with an effects loop that's wired so that the "effects send" jack simply passes the signal from the input stage to the next amplification stage unless there's an effect plugged in. I don't know which amps are wired this way - you'll have to check the schematic on yours to know.] There was a recent discussion thread about bridging the effects loop send jack with a capacitor to reduce the brightness of a Quilter Superblock US. I agree that it's a bit on the bright side with any of the 3 voicings, and I'd be a bit happier with mine if it were a shade darker without compromising its basic character. So I decided to record the same passage with the same guitar through the SBUS with EQ flat and see how and by how much 3 common capacitor values would affect the tone. We're comparing no cap to a 0.047 mF, a 0.15 mF, and a 0.33 mF cap. The only assortment I have is a box of 400V caps, which are much larger than necessary for an input signal. The 400V caps re too big to conceal in the barrel of a phone plug, which is what you'd have to do to use one on a gig.

I used phone plugs with screw terminals (designed to use on speaker wire). These are very handy in the studio.



The test bench is easy to set up. All you have to do to insert a cap is to secure it in a phone plug and plug it in:



I played the first part of My Old Flame as consistently as I could with a D'Andrea ProPlek on my 16" laminated Ibanez AF207 with a Benedetto B7. All EQ on the SBUS was at noon, the volume pot on the guitar was at about 75% and the tone pot was about 60% up (where I seem to get the nicest jazz tone). Here are the sound clips for comparison:

No cap:


0,047 uF:


0.15 uF:


0.33 uF:


The output level did drop by a dB or 2 with each progressive increase in capacitance. You can see this in the peak level (at about 160 Hz) - it drops by about a dB with each increase in capacitance. The overall signal level is even a bit lower than this with each jump because of the loss of highs. Over 0.33 mF, the ouput drops noticeably and could affect gig volume through a small, inefficient speaker. Lest anyone think we're hearing the emperor's new tone spectrum, I also ran spectral analysis on the clips, and here are the results:



So the cap trick works. I may leave the 0.047 in there for the next gig and bring a few others to try as well.

[Stringswinger]

A true gear fanatic would go the extra mile to dial in the tone you want. Bravo David (that is your name right?) for trying this and giving us a detailed report.

I am way too lazy to do any of that. When I feel an amp is too bright (and I felt that way about both my Princeton reissue and my Deluxe reissue), I just move them along. OK, I tried a few mods to the Princeton reissue (speaker swap, solid state rectifier and transformer swap) and it was still too bright for me. I eventually just got a vintage Blackface Princeton and am quite happy with that. But these days I am even too lazy to carry my Princeton (which weighs about 30 pounds with the JBL D-110F that I have in there), so I just bring my Bud 6, a guitar and a cable and I am set. I carry my guitar in a gig bag on one shoulder and the amp in a gig bag on the other shoulder which for a lazy 65 year old (me) is perfect.

[Cunamara]

Everyone will have their preference, for me the .033 cap (correction: now I see it was a 0.33 rather that 0.033 cap) sounded the best of the bunch. I really liked the warmth and solidity it had, especially compared to the amp in its baseline form. In fact, straight from the factory that amp wouldn't last five minutes at my house before it was packed up and going back. The 0.047 and 0.15 didn't correct it enough for me. Interesting to note on the waveform analysis that the 0.33 cap basically chops the top end of the signal off around 5000 Hz, give or take. That's pretty typical for guitar amps, as most guitar speakers don't reproduce anything above that point; the other analyses, and especially that of the unmodified amp, have a lot more very high end content. I think I just don't like hearing those frequencies out of an electric guitar. Does the Toob have an FRFR speaker?

[ccroft]

Nice work Never! You beat me to it. I've been meaning to do a post like this but I'm too lazy, or selfish or whatever.

I made 3 of these cap plugs and fiddled around for a couple of months before deciding I liked the .047 best in my ToneBlock 202. I had a guy put it in the amp with a switch. Now... maybe all of this could've been avoided with a curly chord, but I'll never know :-)

Hey Marc: I'm glad you're as happy with your amp as I am with mine!

chas.

BTW: you can get a wide variety of great sounds out of these Quilters without the cap. But I like the cap just a bit better. It's just a little bit sweeter and lets me vary the treble/mid settings more without it getting too harsh.

[nevershouldhavesoldit]

A true gear fanatic would go the extra mile to dial in the tone you want. Bravo David (that is your name right?) for trying this and giving us a detailed report.

I came close, but stopped short. I want to put a 6 position rotary switch in a small project box, mount 6 caps increasing from 0.047 to 0.55, and plug it into the effects send. Then I could dial in the tone I want easily for any guitar and speaker. I’m thinking of calling it a Varitone switch. Catchy name, isn’t it??

David

[John A.]

Wow, very noticeable difference, even between the lowest value and no cap. If you twiddle knobs on the amp and guitar can you replicate the cap sound without a cap, or is the cap a sound that the amp cannot get otherwise?

[Gitterbug]

Does the Toob have an FRFR speaker?

Hi and thanks for asking! I've made a few 12D models; basically a 10" bass Toob but loaded with Jensen's 150W 12D semi-FRFR speaker. Also, there's the Metro 6.5FR housing 130W SICA 6D 1.5SL dual-cone speaker. Its bass end depends on cab architecture; the newly introduced Metro 6.5 FR II now handles even bass on moderate gig volumes, while extending to 15 kHz. Used for electric & acoustic guitar, manouche, violin, cello, oud, mandolin, harmonica, accordion, keyboard, vocals etc.

[GuyBoden]

A great experiment, the problem I hear is that the higher treble notes are lower in volume than the bass notes as the capacitor values increase.

[0zoro]

Great demo! I think I appreciated the 0.33 cap best, though my ears have become less sensitive with time. I wouldn't mind doing the same with my TB 202. Do you have a link to a more explicit instruction on making the plug? Much appreciated! 0zoro

[nevershouldhavesoldit]

Wow, very noticeable difference, even between the lowest value and no cap. If you twiddle knobs on the amp and guitar can you replicate the cap sound without a cap, or is the cap a sound that the amp cannot get otherwise?

The cap isn't the only element in the filter - there's also resistance from the pots, wiring etc. So even though we're only adding a capacitor to the circuit, we've created and inserted yet another RC filter by doing so. There are several such "minifilters" formed by interaction among the volume and tone pots, tone caps, wiring etc. Adding one more doesn't produce a simple change in the response curve. There are blips and nonlinearities, so that each one has its own effect in the overall circuit.

The EQ pots in the amp have different effective frequency ranges, slopes, cutoffs etc from the added capacitor, so they affect tone differently. I haven't found a combination of settings that did the same thing as the added cap for the SBUS. You could probably come close to the same effect as the cap trick with multiband parametric EQ and/or a variable Q filter set - but a cap is a lot cheaper and easier.

Even a simple single cap RC filter does not introduce a pure and simple 6 dB/octave attenuation curve starting sharply at a single frequency. The actual effect on frequency response is more complex because it's interacting with the circuit. Here's a graph of the kind of real world change a simple filter makes by itself:


I was a bit skeptical of this as a practical improvement I could use - but it is and I will. Keep in mind that the lowest value I used was 0.047 mF. Although standard in Gibsons for many years, this is higher than the tone caps in many modern guitars. I tink the most common value today is 0.022 and it's not uncommon to find caps as low as 0.01 mF.

[Tal_175]

A great experiment, the problem I hear is that the higher treble notes are lower in volume than the bass notes as the capacitor values increase.

Interestingly, based on the frequency graphs, the fundamentals of the high notes don't decrease in volume. The 20th fret of the high E string (C) has the fundamental frequency of about 1000hz. If you look at the graphs, the volume at that frequency changes very little in the first three graphs (stays in the middle of -36 and -42). The perceived volume difference might be due to the reduction of volume in the lower overtones for the treble notes.

[nevershouldhavesoldit]

Interestingly, based on the frequency graphs, the fundamentals of the high notes don't decrease in volume. The 20th fret of the high E string (C) has the fundamental frequency of about 1000hz. If you look at the graphs, the volume at that frequency changes very little in the first three graphs (stays in the middle of -36 and -42). The perceived volume difference might be due to the reduction of volume in the lower overtones for the treble notes.

Look at the areas between 200 and 600 Hz. There’s a clear progressive decline of 1 to 2 dB between caps, which is audible as a change in tone as well as in volume. That little double hump tilts more to the right as more capacitance is added.

[John A.]

The cap isn't the only element in the filter - there's also resistance from the pots, wiring etc. So even though we're only adding a capacitor to the circuit, we've created and inserted yet another RC filter by doing so. There are several such "minifilters" formed by interaction among the volume and tone pots, tone caps, wiring etc. Adding one more doesn't produce a simple change in the response curve. There are blips and nonlinearities, so that each one has its own effect in the overall circuit.

The EQ pots in the amp have different effective frequency ranges, slopes, cutoffs etc from the added capacitor, so they affect tone differently. I haven't found a combination of settings that did the same thing as the added cap for the SBUS. You could probably come close to the same effect as the cap trick with multiband parametric EQ and/or a variable Q filter set - but a cap is a lot cheaper and easier.

Even a simple single cap RC filter does not introduce a pure and simple 6 dB/octave attenuation curve starting sharply at a single frequency. The actual effect on frequency response is more complex because it's interacting with the circuit. Here's a graph of the kind of real world change a simple filter makes by itself:


I was a bit skeptical of this as a practical improvement I could use - but it is and I will. Keep in mind that the lowest value I used was 0.047 mF. Although standard in Gibsons for many years, this is higher than the tone caps in many modern guitars. I tink the most common value today is 0.022 and it's not uncommon to find caps as low as 0.01 mF.

Is that a “yes" or a "no"?

[nevershouldhavesoldit]

Is that a “yes" or a "no"?

”I haven't found a combination of settings that did the same thing as the added cap for the SBUS.”

[GuyBoden]

Look at the areas between 200 and 600 Hz. There’s a clear progressive decline of 1 to 2 dB between caps, which is audible as a change in tone as well as in volume. That little double hump tilts more to the right as more capacitance is added.

I'm wondering if this similar to the tone control on a guitar, when the more you increase Tone control, the less volume there is from two high treble strings?

[John A.]

”I haven't found a combination of settings that did the same thing as the added cap for the SBUS.”

Thanks!

[omphalopsychos]

Nice! Although I had already confirmed this on my own, I didn’t have the patience or time to plot the spectrograms or record multiple clips. Always appreciate your efforts to go above and beyond for the edification of others.

Ever since discovering this, I always use the .047 cap with my superblock.

[ccroft]

”I haven't found a combination of settings that did the same thing as the added cap for the SBUS.”

Same here. Note it's a different Quilter amp though, and I ended up liking the lowest value in Dave's test the best. I still roll back tone a tad on guitar sometimes. I'm also partial to the FRFR voicing these days. That in itself tames the highs a bit.

I'm wondering if this similar to the tone control on a guitar, when the more you increase Tone control, the less volume there is from two high treble strings?

In my comparatively unscientific experiments I find it similar, but not quite the same. In my case that could be placebo effect, but who cares? Might be asking too much of Dave to produce some more spectral analysis :-)

[ccroft]

Great demo! I think I appreciated the 0.33 cap best, though my ears have become less sensitive with time. I wouldn't mind doing the same with my TB 202. Do you have a link to a more explicit instruction on making the plug? Much appreciated! 0zoro

If you've ever repaired a guitar cord you already know how. You just put a capacitor on the plug instead of a cord. One cap lead goes to tip and the other to sleeve. Doesn't matter which one where, and as far as I can tell voltage rating doesn't matter either. There's pictures of other variations towards the end of the thread linked in the first post by NeverShoulda.

I'd suggest trying some of the lower values. It's all personal taste. For me, even the .1 made too much of a difference in volume. This isn't apparent in the sound clips, but it's pretty significant in the room. You can always turn up though. Especially with the TB202.

[nevershouldhavesoldit]

I'm wondering if this similar to the tone control on a guitar, when the more you increase Tone control, the less volume there is from two high treble strings?

It's not that simple because guitars can be wired many different ways. Some behave as you describe and others do not. How the tone pot affects the frequency spectrum of the output signal depends on how the guitar is wired. For example, classic '50s wiring has the tone cap connecting the wiper of the volume pot (the center lug) to lug 3 on the tone pot with the tone pot's wiper grounded. What's often called '60s wiring runs the tone cap from the input lug of the volume pot to lug 3 of the tone pot with the tone pot wiper grounded. And "modern wiring" most often has the tone cap between the input lug on the volume pot and the wiper on the tone pot. With the volume pot fully open, both '50s and modern guitars will sound the same with the same component values. But as soon as you turn down a pot at all, they're different animals.

• With vintage wiring, the tone stays almost the same throughout the full range of the volume pot - treble loss at lower volume settings is greatly reduced. The tone circuit is uncoupled from the volume pot when the volume is below max. So the pickup is uncoupled from the tone circuit and its inductance doesn't interact to change the Q and resonant frequencies of the tone circuit as the tone pot is rotated. But the tone and volume pots do interact with each other. Changing the tone pot affects volume a bit and changing the volume pot affects tone a little bit. Both tone and volume pots produce smoother, more consistent change with rotation, without the minor bumps and dips that occur when pickup inductance is a factor. This particular setup is better for driving amps that clean up with lower guitar volume pot settings.
• With both '60s and modern wiring, the tone pot does not affect the volume. But turning down the volume pot reduces the highs as it cuts volume, so it darkens the tone. There are ways around this too, e.g. what's called a "treble bleed network" (Google it if you're interested).
• The '60s style wiring is functionally identical to modern wiring except for the fact that the grounded tone pot wiper reduces noise from electromagnetic interference. So a guitar with '60s wiring will behave the same (highs drop progressively with volume reduction) but be less prone to noise from RFI and stray EMF.

Here's a simple graphic of this:



There are many ways to wire a guitar. I've even seen a few people put caps of different values from the wiper lug and from one end lug of a tone pot to ground with the wiper and the lug fed by the volume pot shunted, like this -



Some like the tone cap on the volume pot and some put a cap on both volume and tone pots. There's no right or wrong, but every method has pros and cons. Stock wiring may not produce the perfect volume and tone responses for you, and there are many options if you want to pursue your ideal. This is also dependent on the pickup(s) being used, so finding your tone can become a serious effort if you can't find what you want in guitars with one of the traditional wiring schema.

[Herbie]

Wowza, THIS is science!

I have been a bit sceptic about this trick, I have not felt that my Tone Block 202 with a 12" Celestion Cream Alnico is too bright. And if it is, the treble knob has tamed it.

But of course I had to check if I had some extra cap and plug. I had a .047 and a Squareplug. Now I have played with them at home, at rehearsals and on a gig. Not bad.

The question in this subject is to balance with being heard thru the band without getting the 1. and 2. strings too thin. If I adjust the amp too thick, the guitar does not cut thru. I think that this 'cap trick' might help in this quest.

[jzucker]

what speaker is in your toob? Honestly, it sounds darker than my setup even without the capacitor. I suspect your speaker is rolling off a lot of highs. Through a jensen jet tornado, dvmark pearl or ev12L the high freq on my quilter 202 was piercingly bright.

Sadly, while I love the dvmark pearl, it and the quilter together are way too bright for me.

[nevershouldhavesoldit]

what speaker is in your toob? Honestly, it sounds darker than my setup even without the capacitor.

I’m pretty sure it’s a SICA 6L 1.5SL. Hopefully, gitterbug will remember what he put in mine and confirm or correct this.

That particular SICA drops off significantly above about 4500 Hz. At 5kHz, it’s 10 dB below the level from 80 to 4kHz (in which range it’s quite linear for a musical instrument speaker), and it drops continuously from there. Here’s the manufacturer’s frequency response graph for it in a simple vented test cab (which shouldn’t affect the high end at all) -

[jzucker]

pretty dark speaker. Compare that to the emi tonker which drops at about 100db at 6k.

Sica lists it at 60hz-4k. At 4k it's at 90db which is already pretty low.

I think we'd hear a bigger difference in cap values with a brighter speaker.

https://cdn.shopify.com/s/files/1/02...les/Tonker.png

[nevershouldhavesoldit]

pretty dark speaker

Yep! It’s great for that old school electric archtop tone, and it does a nice Claptonesque “woman tone”. But it could use a whizzer cone for those who want a lot more acoustic air and wood in their tone. The Henriksen Blu / Bud 6 has a switchable tweeter for this reason. I like my tone a bit darker and have never used the tweeter. So my Metros are great for my style and use. I have the same SICA in a BG and a GP.

Sica lists it at 60hz-4k. At 4k it's at 90db which is already pretty low……Compare that to the emi tonker which drops at about 100db at 6k.

That’s a separate issue. The little SICA is less sensitive than the EMI, assuming the graph you linked shows the output at 1 meter from a 1 watt input. But it handles a lot of power and makes giggable SPLs even with amps like the Superblock. I’ve used mine with my DV Mark EG250, and it makes a lot of sound.

That EMI has a very audible peak from about 1.6kHz to 5kHz after an equally audible dip from 1 to 1.6kHz. It probably sounds quite bright in any enclosure.