Tube Screamer Mod
Mr. Green
together a simple low gain 5W 1x12ax7, 1x6v6, tube amp. The Tube
Screamer never put out enough volts to overdrive this amp very well.
Thought I'd try LED's in the clipping section of the pedal and the
results were just what I was looking for.
I add a three way switch to give me the original clipping diodes, a
pair of red LEDs or a red LED paired with two rectifier diodes in
series. The original diodes clipped at about 0.6V so no matter how
high you turned up the volume, that was the most you could ever get
out of the pedal. The LEDs clip at 1.6V while the two rectifier diodes
in series clip at about 1.3V. These higher voltages really drive my
valve amp well.
Tone wise the LEDs make the pedal sound much clearer. This is probably
partly down to the fact that you're adding more gain to the
frequencies the pedal is designed to clip and those are highs.
If your amp needs a bit more of a push I definitely recommend giving
this mod a try. It's pretty easy to do and makes a big difference. OK
it will not turn a clean amp into a metal machine but it's great for a
good blues/rock overdrive.
Of course changing the diodes, in any overdrive pedal, to LEDs should
give you a similar increase in output.
MrGreen
Rocket Scientist
<cl...@wheatleymetalfabrications.co.uk> wrote:
>A while back now I built a Tube Screamer Clone and before that I put
>together a simple low gain 5W 1x12ax7, 1x6v6, tube amp. The Tube
>Screamer never put out enough volts to overdrive this amp very well.
>Thought I'd try LED's in the clipping section of the pedal and the
>results were just what I was looking for.
I'm surprised that the original TubeScreamer circuit wasn't driving
your amp well, at least as compared to clean signal. Maybe that's due
to the single 12ax7, but that should also be the case when the TS is
bypassed.
>I add a three way switch to give me the original clipping diodes, a
>pair of red LEDs or a red LED paired with two rectifier diodes in
>series. The original diodes clipped at about 0.6V so no matter how
>high you turned up the volume, that was the most you could ever get
>out of the pedal. The LEDs clip at 1.6V while the two rectifier diodes
>in series clip at about 1.3V. These higher voltages really drive my
>valve amp well.
>
>Tone wise the LEDs make the pedal sound much clearer. This is probably
>partly down to the fact that you're adding more gain to the
>frequencies the pedal is designed to clip and those are highs.
That would be due to the higher clipping threshold. The internal
pre-drive (clean) stage of the TS has not been altered, so you have
proportionately less drive compared to the 'clip' level of the diodes.
(Since the diode clip level is increased)
>If your amp needs a bit more of a push I definitely recommend giving
>this mod a try. It's pretty easy to do and makes a big difference. OK
>it will not turn a clean amp into a metal machine but it's great for a
>good blues/rock overdrive.
>
>Of course changing the diodes, in any overdrive pedal, to LEDs should
>give you a similar increase in output.
>
>MrGreen
Different types or colors of LEDs will also give different break-over
voltages. Good post, Mr. Green.
WB
>
> Different types or colors of LEDs will also give different break-over
> voltages. Good post, Mr. Green.
Clipping occurs on diodes ( aka LEDs too ) when the junction voltage
exceeds approx. .7vdc . Thus by stacking diodes in series across
the output - you can alter the clipping level.
Jim
> On Tue, 21 Dec 2010 05:18:14 -0800 (PST), "Mr. Green"
> <cl...@wheatleymetalfabrications.co.uk> wrote:
>
>> A while back now I built a Tube Screamer Clone and before that I put
>> together a simple low gain 5W 1x12ax7, 1x6v6, tube amp. The Tube
>> Screamer never put out enough volts to overdrive this amp very well.
>> Thought I'd try LED's in the clipping section of the pedal and the
>> results were just what I was looking for.
>
> I'm surprised that the original TubeScreamer circuit wasn't driving
> your amp well, at least as compared to clean signal. Maybe that's due
> to the single 12ax7, but that should also be the case when the TS is
> bypassed.
Sounds like he wanted a CLEANER TS with the drive down. Even with the
drive down, the TS isn't really clean. Not that you'd notice that much,
once you overdrive the amp!
>
>> I add a three way switch to give me the original clipping diodes, a
>> pair of red LEDs or a red LED paired with two rectifier diodes in
>> series. The original diodes clipped at about 0.6V so no matter how
>> high you turned up the volume, that was the most you could ever get
>> out of the pedal. The LEDs clip at 1.6V while the two rectifier diodes
>> in series clip at about 1.3V. These higher voltages really drive my
>> valve amp well.
>>
>> Tone wise the LEDs make the pedal sound much clearer. This is probably
>> partly down to the fact that you're adding more gain to the
>> frequencies the pedal is designed to clip and those are highs.
>
> That would be due to the higher clipping threshold. The internal
> pre-drive (clean) stage of the TS has not been altered, so you have
> proportionately less drive compared to the 'clip' level of the diodes.
> (Since the diode clip level is increased)
>
>> If your amp needs a bit more of a push I definitely recommend giving
>> this mod a try. It's pretty easy to do and makes a big difference. OK
>> it will not turn a clean amp into a metal machine but it's great for a
>> good blues/rock overdrive.
>>
>> Of course changing the diodes, in any overdrive pedal, to LEDs should
>> give you a similar increase in output.
>>
>> MrGreen
>
> Different types or colors of LEDs will also give different break-over
> voltages. Good post, Mr. Green.
I even sort my LED's by threshold voltage within color. I used
different color LED's in one of my mods, with the widest voltage
variation I could find, for a subtle but noticeable asymmetrical clipping.
Rocket Scientist
The 0.65 to 0.70 v number is valid for silicon diodes. LEDs have a
higher breakover voltage, and it's dependent on the color/manufacturer
of the LED. Germanium diodes break over at about 0.2 volts, so you'd
get more distortion but lower output. At an extreme, there are
"Schottky diodes" which have an even lower breakover voltage.
Yes, you can put multiple diodes in series to get higher breakover
voltages. The OP already mentioned that, so I didn't reiterate, but
it's good for when you like the sound of a certain type of diode but
you want higher output and less overdrive (they go hand in hand unless
the pre-drive voltage stage is altered).
I personally like the sound of 1N400x-series rectiier diodes, where
the X is a number, like 1N4001, 1N4004. That last number designates
the peak -reverse- voltage, which doesn't matter in this case. You're
OK with whatever Radio Shack has.
Another slight mod, if you want to try it: Put a small value cap
across (parallel to) the diodes. Just one will do, as they're not
directional. Depending on value, that will round off some highs. Of
course you wouldn't do that if you think your OD circuit is dull
sounding to begin with.
RS
Jim
> On Tue, 21 Dec 2010 11:36:41 -0600, WB<nos...@no.com> wrote:
>
>> On 12/21/2010 11:10 AM, Rocket Scientist wrote
>>>
>>> Different types or colors of LEDs will also give different break-over
>>> voltages. Good post, Mr. Green.
>>
>> Clipping occurs on diodes ( aka LEDs too ) when the junction voltage
>> exceeds approx. .7vdc . Thus by stacking diodes in series across
>> the output - you can alter the clipping level.
>
> The 0.65 to 0.70 v number is valid for silicon diodes. LEDs have a
> higher breakover voltage, and it's dependent on the color/manufacturer
> of the LED. Germanium diodes break over at about 0.2 volts, so you'd
> get more distortion but lower output. At an extreme, there are
> "Schottky diodes" which have an even lower breakover voltage.
>
> Yes, you can put multiple diodes in series to get higher breakover
> voltages. The OP already mentioned that, so I didn't reiterate, but
> it's good for when you like the sound of a certain type of diode but
> you want higher output and less overdrive (they go hand in hand unless
> the pre-drive voltage stage is altered).
>
> I personally like the sound of 1N400x-series rectiier diodes, where
> the X is a number, like 1N4001, 1N4004. That last number designates
> the peak -reverse- voltage, which doesn't matter in this case. You're
> OK with whatever Radio Shack has.
I do the stacked tiny diodes and different color LEDs, but I'll try
rectifier diodes the next time around. I have several in my amp
repair/building stock.
I also recommend experimenting with clipping at different stages. In
the loop, as with the TS-9/808, and after the gain stage.
>
> Another slight mod, if you want to try it: Put a small value cap
> across (parallel to) the diodes. Just one will do, as they're not
> directional. Depending on value, that will round off some highs.
I always experiment with that cap to calm the typical "wasp in a
Folger's can" buzz of most of these circuits.
Tony Done
"Mr. Green" <cl...@wheatleymetalfabrications.co.uk> wrote in message
news:fc088aa3-4bfc-4e60...@w2g2000yqb.googlegroups.com...
Interesting post, thanks. I did a switchable asymmetric clipping mode on a
TS-type pedal - Boss BD-2, because I preferred the sound of the of the
(asymmetric) Boss SD-1. Sure enough I liked the asymmetric option better on
the BD-2.
For anyone interested in very wide ranges of gain, OD and tone, the Sansamp
pedals seem to have it. I have the Character Series "British", and I'm think
of getting their Para Acoustic for versatile clean gain.
Tony D
Meat Plow
Try some Schottky diodes!
--
Live Fast, Die Young and Leave a Pretty Corpse
WB
> Thus by stacking diodes in series across the
>> output - you can alter the clipping level.
>
> Try some Schottky diodes!
>
>
>
Mr. Green
>
> > I'm surprised that the original TubeScreamer circuit wasn't driving
> > your amp well, at least as compared to clean signal. Maybe that's due
> > to the single 12ax7, but that should also be the case when the TS is
> > bypassed.
>
> Sounds like he wanted a CLEANER TS with the drive down. Even with the
> drive down, the TS isn't really clean. Not that you'd notice that much,
> once you overdrive the amp!
>
The amp only has a single 12ax7 so not a lot of gain however, I did
build it with a seperate gain and master volume so I can still drive
the preamp at lower volumes. Yes I wanted to drive the pre amp with a
fairly clean signal from the TubeScreamer. That way I get the fatter
tone from the TS with clipping mainly from the amp. Ofcourse using a
standard TS, even with the gain maxed you don't get more than about
0.6V out of it.
The cleaner / brighter tone, with the LED's, suits my teles neck
pickup nicely but my favourite switch position seems to be the
slightly asymmetrical option of one red LED allowing the current to
flow one way and, a pair of rectifier diodes in series going the other
way. That gives 1.6V against 1.3V, lots of output but a bit warmer
tone than the pair of LEDs.
As you said, even with the drive down an standard TS still clips. If
someones looking to hear more of just the valve amp clipping I think
this mod does the trick well. I wasn't after loads of distortion. I
wanted a great just breaking-up sound. Now my TS is just shaping the
tone and boosting my guitar signal while, valves in the amp are doing
what they do best.
Thanks for you're interest, MrGreen
Jim
> On Dec 21, 6:16 pm, Jim<inse@ttle> wrote:
> >
>>> I'm surprised that the original TubeScreamer circuit wasn't driving
>>> your amp well, at least as compared to clean signal. Maybe that's due
>>> to the single 12ax7, but that should also be the case when the TS is
>>> bypassed.
>>
>> Sounds like he wanted a CLEANER TS with the drive down. Even with the
>> drive down, the TS isn't really clean. Not that you'd notice that much,
>> once you overdrive the amp!
>>
>
> The amp only has a single 12ax7 so not a lot of gain however, I did
> build it with a seperate gain and master volume so I can still drive
> the preamp at lower volumes. Yes I wanted to drive the pre amp with a
> fairly clean signal from the TubeScreamer. That way I get the fatter
> tone from the TS with clipping mainly from the amp. Ofcourse using a
> standard TS, even with the gain maxed you don't get more than about
> 0.6V out of it.
Out of curiosity, I just tested my original TS-9 modified to 808. I
used a signal generator, a quality true RMS Fluke meter, and a scope.
I *can* get over 1V out, but that's with a bit over 300 mV in. At 100
mV in, I'm around your 600 mV figure. And it didn't look linear. Lower
input levels looked like they were giving higher gain factors. NOTE:
In the previous sentence I use "gain" in the technical sense --
amplification factor, not distortion tone!
And as our ears tell us, it's not clean. Waveform varies according to
frequency. At a 1 kHz common test frequency, it's not that dirty
looking (close to a sine wave). But at 300 Hz, it's pretty ugly, even
with the drive down (and level UP). Also, I could see how the tone
control works. What's very interesting to me is that I tend to set it
right before it starts showing a big increase in clipping (which is
fairly high, over 3:00). As you reduce the tone, it starts looking like
a saw tooth waveform.
This was quick and dirty. If there's some interest from a few folks, I
could set up a tripod and snap some pics of the scope screen.
I also tossed on my DOD 250 early gray circuit build (on perf board, in
a reissue 250 case). Surprisingly, it looked like it had less
amplification factor (technical gain). It is cleaner with the "gain"
down, but it'll go full-on square wave with gain up. I've been planning
to add an asymmetrical mod to this one, so I think I'll leave it sitting
on my bench.
The other two I tossed on were very radical type pedals. A stock DOD
Death Metal (platform for my mods), and a Rocktron Zombie. Both show
heavy distortion with lots of harmonics. And the Death Metal showed a
very reactive tone control section (why I use it for mods).
The last thing I did was put my EMG 81/85 equipped SG on the Fluke 77
A/N. This has obvious flaws, but it put out averages of around 300 mV
with fairly heavy open strums. But when I turned up the SPC active
tone? Over 1V! Now I know why I like to use the SPC boost with tube amps.
Enough rambling for today...
Mr. Green
> On 12/22/2010 1:49 AM, Mr. Green wrote:
>
>
>
>
>
> > On Dec 21, 6:16 pm, Jim<inse@ttle> wrote:
>
> >>> I'm surprised that the original TubeScreamer circuit wasn't driving
> >>> your amp well, at least as compared to clean signal. Maybe that's due
> >>> to the single 12ax7, but that should also be the case when the TS is
> >>> bypassed.
>
> >> Sounds like he wanted a CLEANER TS with the drive down. Even with the
> >> drive down, the TS isn't really clean. Not that you'd notice that much,
> >> once you overdrive the amp!
>
> > The amp only has a single 12ax7 so not a lot of gain however, I did
> > build it with a seperate gain and master volume so I can still drive
> > the preamp at lower volumes. Yes I wanted to drive the pre amp with a
> > fairly clean signal from the TubeScreamer. That way I get the fatter
> > tone from the TS with clipping mainly from the amp. Ofcourse using a
> > standard TS, even with the gain maxed you don't get more than about
> > 0.6V out of it.
>
> Out of curiosity, I just tested my original TS-9 modified to 808. I
> used a signal generator, a quality true RMS Fluke meter, and a scope.
>
> I *can* get over 1V out, but that's with a bit over 300 mV in. At 100
> mV in, I'm around your 600 mV figure. And it didn't look linear. Lower
> input levels looked like they were giving higher gain factors. NOTE:
> In the previous sentence I use "gain" in the technical sense --
> amplification factor, not distortion tone!
Thanks for posting this, it's really interesting stuff. I don't have
anything more than a cheap multimeter.
Hadn't thought of this before but the signal could be getting a bit of
a boost from the active tone section after the clipping stage. That
may explain how you can get over the expected 0.6V output. What do you
think. Be interesting to see how much the tone control effects the
output.
>
> And as our ears tell us, it's not clean. Waveform varies according to
> frequency. At a 1 kHz common test frequency, it's not that dirty
> looking (close to a sine wave). But at 300 Hz, it's pretty ugly, even
> with the drive down (and level UP). Also, I could see how the tone
> control works. What's very interesting to me is that I tend to set it
> right before it starts showing a big increase in clipping (which is
> fairly high, over 3:00). As you reduce the tone, it starts looking like
> a saw tooth waveform.
Ah you did play with the tone control. So the lower frequencies are
more distorted. That's interesting 'cause I was under the impression
TS's sounded smoother because the distortion was limited more to the
higher frequencies. Could it be that the chip lets through more
undistorted highs in the clean element of the signal?
>
> This was quick and dirty. If there's some interest from a few folks, I
> could set up a tripod and snap some pics of the scope screen.
>
> I also tossed on my DOD 250 early gray circuit build (on perf board, in
> a reissue 250 case). Surprisingly, it looked like it had less
> amplification factor (technical gain). It is cleaner with the "gain"
> down, but it'll go full-on square wave with gain up. I've been planning
> to add an asymmetrical mod to this one, so I think I'll leave it sitting
> on my bench.
So the TS distortion does look smoother compared to the DOD.
>
> The other two I tossed on were very radical type pedals. A stock DOD
> Death Metal (platform for my mods), and a Rocktron Zombie. Both show
> heavy distortion with lots of harmonics. And the Death Metal showed a
> very reactive tone control section (why I use it for mods).
>
> The last thing I did was put my EMG 81/85 equipped SG on the Fluke 77
> A/N. This has obvious flaws, but it put out averages of around 300 mV
> with fairly heavy open strums. But when I turned up the SPC active
> tone? Over 1V! Now I know why I like to use the SPC boost with tube amps.
I believe tube stages are bias at about 1.3V so, over 1V will max out
the first preamp stage which means, with the volume up, you've got to
be getting clipping from the second stage. At least that's how I see
it ;-)
>
> Enough rambling for today...- Hide quoted text -
>
> - Show quoted text -
Thanks again, good post.
MrGreen
Mr. Green
> > On 12/22/2010 1:49 AM, Mr. Green wrote:
> >> The amp only has a single 12ax7 so not a lot of gain however, I did
> >> build it with a seperate gain and master volume so I can still drive
> >> the preamp at lower volumes. Yes I wanted to drive the pre amp with a
> >> fairly clean signal from the TubeScreamer. That way I get the fatter
> >> tone from the TS with clipping mainly from the amp. Ofcourse using a
> >> standard TS, even with the gain maxed you don't get more than about
> >> 0.6V out of it.
>
>
> - Silicon
> - LED
> - Bypass diodes altogether
>
> Bypassing them altogether gives the biggest 'clean' boost, but the
> silicon diodes actually sound the 'nicest' out of the three options.
>
> This is the same as a certain very expensive clone which sells for
> around $400 :-)
>
> --
> Usenet Improvement Project:http://twovoyagers.com/improve-usenet.org/
I've seen an expensive clone with that switch combination. I used a
three way switch (on, off, on toggle) and wired the pair of LEDs
across the centre terminal so they are always in circuit. I put the
original pair of silicon signal diode in one "on" position and, two
rectifier diodes in series in the other "on" position. I didn't fancy
the no diode option. As the signal diodes clip well before the LEDs,
it doesn't that the LEDs are always in circuit.
I think different switch positions suit different music. The LEDs give
great low gain clarity which allows you to still hear the individual
notes in a strummed chord. The LED / rectifier diode combination is
clear but a bit warmer so suits low gain lead playing and crunchy rock
chords. The original signal diodes are probably the smoothest sounding
with more distortion available from the pedal into a clean amp. I like
them all but the new options give me what I was looking for from my
amp.
Have you built many pedals? I've only put together overdrive units,
simple boosters and buffers. I was interested in making a tremelo
pedal, ever tried making one?
MrGreen
Les Cargill
> On 12/22/2010 1:49 AM, Mr. Green wrote:
>> On Dec 21, 6:16 pm, Jim<inse@ttle> wrote:
>> >
>>>> I'm surprised that the original TubeScreamer circuit wasn't driving
>>>> your amp well, at least as compared to clean signal. Maybe that's due
>>>> to the single 12ax7, but that should also be the case when the TS is
>>>> bypassed.
>>>
>>> Sounds like he wanted a CLEANER TS with the drive down. Even with the
>>> drive down, the TS isn't really clean. Not that you'd notice that much,
>>> once you overdrive the amp!
>>>
>>
>> The amp only has a single 12ax7 so not a lot of gain however, I did
>> build it with a seperate gain and master volume so I can still drive
>> the preamp at lower volumes. Yes I wanted to drive the pre amp with a
>> fairly clean signal from the TubeScreamer. That way I get the fatter
>> tone from the TS with clipping mainly from the amp. Ofcourse using a
>> standard TS, even with the gain maxed you don't get more than about
>> 0.6V out of it.
>
> Out of curiosity, I just tested my original TS-9 modified to 808. I used
> a signal generator, a quality true RMS Fluke meter, and a scope.
>
> I *can* get over 1V out, but that's with a bit over 300 mV in. At 100 mV
> in, I'm around your 600 mV figure. And it didn't look linear.
It most likely acts as a limiter of unusual slope ( I am guessing ).
(assuming this schem is relevant...)
http://www.geofex.com/article_folders/TStech/tsxtech.gif
Most of the "fun" is in the feedback leg of A1... not
exactly a classic limiter, but ...
That 51p cap in shunt across the feedback leg, in parallel
with those diodes... oy. I presume it's a gain-based
lowpass filter. But you won't get a nice consistent
filter out of it - which is probably the point...
Pedals are weird.
> Lower
> input levels looked like they were giving higher gain factors. NOTE: In
> the previous sentence I use "gain" in the technical sense --
> amplification factor, not distortion tone!
>
> And as our ears tell us, it's not clean. Waveform varies according to
> frequency. At a 1 kHz common test frequency, it's not that dirty looking
> (close to a sine wave). But at 300 Hz, it's pretty ugly, even with the
> drive down (and level UP). Also, I could see how the tone control works.
> What's very interesting to me is that I tend to set it right before it
> starts showing a big increase in clipping (which is fairly high, over
> 3:00). As you reduce the tone, it starts looking like a saw tooth waveform.
>
> This was quick and dirty. If there's some interest from a few folks, I
> could set up a tripod and snap some pics of the scope screen.
>
> I also tossed on my DOD 250 early gray circuit build (on perf board, in
> a reissue 250 case). Surprisingly, it looked like it had less
> amplification factor (technical gain). It is cleaner with the "gain"
> down, but it'll go full-on square wave with gain up. I've been planning
> to add an asymmetrical mod to this one, so I think I'll leave it sitting
> on my bench.
>
Asymmetric is very strange. I've worked with that a bit in the digital
domain. It's ... "organic" ( meaning it's kind of what you expect from
a Champ in distress ) ... ish. Pushes you into Neil Young territory.
> The other two I tossed on were very radical type pedals. A stock DOD
> Death Metal (platform for my mods), and a Rocktron Zombie. Both show
> heavy distortion with lots of harmonics. And the Death Metal showed a
> very reactive tone control section (why I use it for mods).
>
> The last thing I did was put my EMG 81/85 equipped SG on the Fluke 77
> A/N. This has obvious flaws, but it put out averages of around 300 mV
> with fairly heavy open strums. But when I turned up the SPC active tone?
> Over 1V! Now I know why I like to use the SPC boost with tube amps.
>
99% of "it sounds better" is "it's louder" with guitar stuff.
> Enough rambling for today...
Nice poast.
--
Les Cargill
Rocket Scientist
<cl...@wheatleymetalfabrications.co.uk> wrote:
>Have you built many pedals? I've only put together overdrive units,
>simple boosters and buffers. I was interested in making a tremelo
>pedal, ever tried making one?
>
>MrGreen
I've designed and built circuits in that domain, including panners,
phasers and flangers. Never had much interest in tremolo, but the same
concepts apply.
The shape of the modulation waveform can be crucial. Some schemes
yield relatively smooth motion on the 'top' side of the mod waveform,
but then a sharp dip as it goes to low resistance (min volume in the
case of the trem circuit). That's the main thing to pay attention to.
If you have a chance to audition a type of pedal before building it,
listen for that. I've sometimes done waveshaping on the modulation
waveform just to get that to sound smooth.
There are also several different methods for modulating the audio:
FET/Bipolar, Optical/Cadmium Sulfide (as on Fender amps), OTA (a type
of opamp). Each has its own quirks.
There's a site called "BuildYourOwnClone.com" that has kits, if
you're interested. Their trem is FET/Bipolar. I haven't heard it, so I
can't comment on the sound.
http://www.buildyourownclone.com/tremolo.html
Rocket Scientist
<cl...@wheatleymetalfabrications.co.uk> wrote:
>On Dec 22, 9:31 pm, Jim <inse@ttle> wrote:
>> I *can* get over 1V out, but that's with a bit over 300 mV in. At 100
>> mV in, I'm around your 600 mV figure. And it didn't look linear. Lower
>> input levels looked like they were giving higher gain factors. NOTE:
>> In the previous sentence I use "gain" in the technical sense --
>> amplification factor, not distortion tone!
>
>Thanks for posting this, it's really interesting stuff. I don't have
>anything more than a cheap multimeter.
>Hadn't thought of this before but the signal could be getting a bit of
>a boost from the active tone section after the clipping stage. That
>may explain how you can get over the expected 0.6V output. What do you
>think. Be interesting to see how much the tone control effects the
>output.
The diodes limit the + and - separately, so you get 0.65 from each
side = 1.3v peak to peak.
If I remember the output/tone stage correctly, you're getting near
unity gain. Unless you alter the way the tone control works, you can't
just boost the gain there (as you easily could if it were just a
normal opamp gain stage). That was the first thing that occurred to me
when you mentioned increased output, but I think it's easier to just
use more diodes unless you're looking for extreme crunch -plus-
increased gain.
Les Cargill
> On Thu, 23 Dec 2010 02:28:03 -0800 (PST), "Mr. Green"
> <cl...@wheatleymetalfabrications.co.uk> wrote:
>
>> On Dec 22, 9:31 pm, Jim<inse@ttle> wrote:
>
>>> I *can* get over 1V out, but that's with a bit over 300 mV in. At 100
>>> mV in, I'm around your 600 mV figure. And it didn't look linear. Lower
>>> input levels looked like they were giving higher gain factors. NOTE:
>>> In the previous sentence I use "gain" in the technical sense --
>>> amplification factor, not distortion tone!
>>
>> Thanks for posting this, it's really interesting stuff. I don't have
>> anything more than a cheap multimeter.
>> Hadn't thought of this before but the signal could be getting a bit of
>> a boost from the active tone section after the clipping stage. That
>> may explain how you can get over the expected 0.6V output. What do you
>> think. Be interesting to see how much the tone control effects the
>> output.
>
> The diodes limit the + and - separately, so you get 0.65 from each
> side = 1.3v peak to peak.
>
I imagine that's an asymptopic figure, unless I miss my guess
on how those diodes conduct...
> If I remember the output/tone stage correctly, you're getting near
> unity gain. Unless you alter the way the tone control works, you can't
> just boost the gain there (as you easily could if it were just a
> normal opamp gain stage). That was the first thing that occurred to me
> when you mentioned increased output, but I think it's easier to just
> use more diodes unless you're looking for extreme crunch -plus-
> increased gain.
--
Les Cargill
RS
<lcarg...@comcast.net> wrote:
>Rocket Scientist wrote:
>>
>> The diodes limit the + and - separately, so you get 0.65 from each
>> side = 1.3v peak to peak.
>I imagine that's an asymptopic figure, unless I miss my guess
>on how those diodes conduct...
Yeah, they're more or less logarithmic. The base-emitter junction of a
regular bipolar transistor shows the same kind of breakover
characteristic. In fact, that's another viable overdrive element:
Use the B-E leads of bipolar transistors instead of the diodes.
Back-to-back, of course. You could connect each collector lead to the
base, but that shouldn't matter much in this app.
Rocket Scientist
Oh...and Darlington transistors would give you the same kind of effect
as hooking two diodes in series. Double the voltage out, with
proportionately less overdrive.
Lord Valve
That's "asymptotic." Just thought you college boys would like to know.
After all, your functional intelligence approaches
infinity, at least, to hear you tell it.
Lord Valve
Expert (pleases obsess)
Jim
I just tossed it back on for a minute, this time choosing 250 Hz as a
test frequency.
Measured with a Fluke 77 A/N, Constants: 100 mV at 250 Hz input, drive
set to minimum, level set to maximum.
Tone setting, with output:
Min. = 230 mV
9:00 = 225 mV
12:00 = 270 mV
3:00 = 312 mV
Max. = 297 mV
Also checked bypass, 100mV at 250 Hz in = 95 mV out, and I can't see
any distortion on the scope (so it's pretty close to "unity," just 5% low).
Now, with these constants, I'll look at output...
Constants: 100 mV at 250 Hz in, tone at 3:00, level pegged:
Variable = Drive
Min. = 309 mV
9:00 = 347
12:00 = 384
3:00 = 406
Max. = 409 mV
And this was all done with 100 mV in. To get those 600 mV out that we
were talking about, you need a hotter input signal.
>
>>
>> And as our ears tell us, it's not clean. Waveform varies according to
>> frequency. At a 1 kHz common test frequency, it's not that dirty
>> looking (close to a sine wave). But at 300 Hz, it's pretty ugly, even
>> with the drive down (and level UP). Also, I could see how the tone
>> control works. What's very interesting to me is that I tend to set it
>> right before it starts showing a big increase in clipping (which is
>> fairly high, over 3:00). As you reduce the tone, it starts looking like
>> a saw tooth waveform.
>
> Ah you did play with the tone control. So the lower frequencies are
> more distorted. That's interesting 'cause I was under the impression
> TS's sounded smoother because the distortion was limited more to the
> higher frequencies.
The TS is known for emphasizing mids. So now let's look at keeping
these factors constant: Drive 9:00, Tone 3:00, Level pegged.
We vary the input frequency (at 100 mV), and watch the output voltage...
80 Hz = 202 mV
160 Hz = 279
320 Hz = 383
500 Hz = PEAK OUTPUT 415 mV
640 Hz = 406
1280 Hz = 309
2560 Hz = 147
5120 Hz = 65
The high cut may account for the "more natural, tube-like" reputation of
the TS. The harsher higher harmonics are being attenuated, as well as
the low end (which can make preamp tubes sound mushy).
So at this setting, we're seeing about a 3dB attenuation in the bottom
of the guitar spectrum, but even more in the highest frequency range
(harmonics of the fundamental frequencies).
NOTE: you see variations from one test to another of around 1% in
voltage. I didn't recalibrate the 100 mV input, so suspect that the
fridge or electric heat kicked on, causing a slight line variation. But
hey, 1% is close anyway!
> Could it be that the chip lets through more
> undistorted highs in the clean element of the signal?
"Lows," "mids" and "highs" are all subjective when it comes to guitar.
But you can look at that bare-bones frequency chart above to see what
happens (at 9:00 drive, 3:00 tone).
RS
<detr...@ix.netcom.com> wrote:
>RS wrote:
>
>> On Thu, 23 Dec 2010 13:48:41 -0500, Les Cargill
>> <lcarg...@comcast.net> wrote:
>>
>> >Rocket Scientist wrote:
>> >>
>> >> The diodes limit the + and - separately, so you get 0.65 from each
>> >> side = 1.3v peak to peak.
>>
>> >I imagine that's an asymptopic figure, unless I miss my guess
>> >on how those diodes conduct...
>>
>> Yeah, they're more or less logarithmic. The base-emitter junction of a
>> regular bipolar transistor shows the same kind of breakover
>> characteristic. In fact, that's another viable overdrive element:
>> Use the B-E leads of bipolar transistors instead of the diodes.
>> Back-to-back, of course. You could connect each collector lead to the
>> base, but that shouldn't matter much in this app.
>
>That's "asymptotic." Just thought you college boys would like to know.
...said the guy who misspelled his own name right after calling
someone an idiot. That must have been embarrassing. So, "Lorfd", if I
may call you "Lorfd", you have very little with regard to math chops,
or electronic engineering chops for that matter (which will be easy to
confirm if you want to continue). Why on earth would you jump in to
nitpick Les's obvious typo?
>After all, your functional intelligence approaches
>infinity,...
Understandable that it appears that way to you.
>Lord Valve
>Expert (pleases obsess)
And sure enough, signed off as 'expert' after that. Look up the word
"irony." If you had taken the time to study electronic engineering or
even played a bit of guitar, you might have been able to contribute
something useful.
If you really want to show off your claimed 'expertise', start by
explaining the gain equation and tone shaping curves of the TS
post-drive opamp tone circuit. Yep, time for Lorfd to disappear.
BTW, "pleases obsess"? Don't you hate that, right after correcting
someone else? You always manage to step in your own pile. Run along
now. <g>
Lord Valve
Barrel. Shotgun. <shrug>
Lord Valve
Expert (please osbess)
Rocket Scientist
>
>I just tossed it back on for a minute, this time choosing 250 Hz as a
>test frequency.
>
>Measured with a Fluke 77 A/N, Constants: 100 mV at 250 Hz input, drive
>set to minimum, level set to maximum.
>
>Tone setting, with output:
>
>Min. = 230 mV
>9:00 = 225 mV
>12:00 = 270 mV
>3:00 = 312 mV
>Max. = 297 mV
Jim, Interesting posts. It sounds like you'd have fun with a 'sweep
generator.' I use then when charting frequency response, especially
when working on tone controls. Set the lowest frequency and the
highest frequency. The sweep generator does an incremental sweep
through that range.
The goal is to display the output on a scope screen with the lowest
freq on the left of the trace, highest at right. That requires that
the sweep gen send out a 'ping' (separate output pin) when it starts
the sweep, so you can trigger the horizontal sweep on the scope. If
you get one, make sure it does have that output (some inexplicably do
not).
Also good to be able to switch between linear and log for the sweep
'rate', so you can view the output in roughly the same way as the ear
hears octaves/pitch.
Rocket Scientist
<detr...@ix.netcom.com> wrote:
>RS wrote:
>> If you really want to show off your claimed 'expertise', start by
>> explaining the gain equation and tone shaping curves of the TS
>> post-drive opamp tone circuit. Yep, time for Lorfd to disappear.
>Barrel. Shotgun. <shrug>
>
>Lord Valve
>Expert (please osbess)
??? So you jumped into the pool trying to make a big splash but forgot
you don't know how to swim. Unless your goal was to make it obvious
that you don't really know how opamps work (and you don't), that
would seem counterproductive. Or you could try to prove me wrong and
show the math for the opamp output/tone circuit.
Otherwise, how about keeping your garbage on aga?