Hoffman Amplifiers Tube Amplifier Forum
Other Stuff => Solid State => Topic started by: Fresh_Start on November 22, 2010, 09:21:25 am
-
Can someone please explain to me what is going on with the clipping circuit(s) in this amp?
http://www.drtube.com/schematics/marshall/2550amp.gif
I understand the basic idea behind a diode clipper.
What does the 1nf cap in parallel with the bigger diode network do?
What is the logic behind the diode clipper network? IOW the 2 LEDs connected on both ends above the un-balanced 2-1 clipper.
I've heard several guys I respect a lot say good things about how this circuit works so I'd like to understand it better.
Thanks for any insights,
Chip
-
I'd say that the cap forms a low-pass filter and that the LEDs are used because of their higher forward voltage drop (1.5 V for std models vs. 0.7 V for typical 1N4xxx). So we have > 1.5 V clipped symetrically by the LEDs and > 1.5 V/0.7 V clipped asymetrically by the bottom 1N4xxxs, 3V one way, 2.2 V the other. Right? Well, no, as we need to adjust for forward drop. See http://www.allaboutcircuits.com/vol_3/chpt_3/6.html for a discussion and simulation, where it's shown that D1 will clip at 2 V and D2 at -3 V. Adding a second diode in series somewhat balances the clipper at around ± 3 V.
-
That drawing is a mish-mosh.
RE-DRAW it until it makes less nonsense.
(http://i55.tinypic.com/2qi8zn8.jpg)
There's a gain of about 1,500 (adjustable down 40dB) from the input. The amp output is near 39K, padded-up with 47K and loaded with 220K.
We know that diodes will clip at a few volts. A few more volts across 86K will be a few dozen microamps. Diode drops will be low, maybe 0.1V lower than we assume for 1mA.
I read the diode drops as 1.5V LED, 0.5V 4007. The way they are stacked up gives about 20% asymmetry, 2.0V and 2.5V.
The 1nFd trims the highs. This is not just clip-then-filter, the cap AT the diodes reduces clipping of high harmonics, which would throw hash up the spectrum which would be hard to filter.
There's a 160Hz bass-shelf in the first stage.
No real need to understand. There's no real reason not to copy it as-is.
> Right? Well, no...
You mean this?
(http://sub.allaboutcircuits.com/images/03439.png)
Note that this has arbitrary 1.3V and -2.3V voltage-sources; the Marshall does not. Their answer is correct IF you put those batteries in.
-
How ever you do it, don't forget that coupling cap before the circuit.
-
There are some circuits at work that use back to back series connected 4.7v zeners across an op amp inputs to clip/limit the input signal to approx +5 volts. I've never seen this clipping method used in guitar amps. Is the zener cost a major factor or is there some other reason to not use zeners for guitar clippers?
-
Zeners are quite noisy.
-
> Is the zener cost a major factor
This isn't government work. Even "minor cost" must be avoided.
> or is there some other reason to not use zeners for guitar clippers?
Also there is little need for 5V signals, and Zeners don't come a lot lower.
Noise is probably the least of the reasons. There's no significant noise when not conducting. When conducting, there's so much fuzz-hash, who hears noise? This "conduction noise" is down in the mV, 60dB down from clipping level.
> don't forget that coupling cap before the circuit.
That's a major part, true. The result is always asymmetric, but the asymmetry changes over time.
Which is another reason to PLAGIARIZE rather than analyze. We can say "Oh, it clips!" but we can't say "It sounds keen because _______" until we work with it, and maybe not even then. Since it basically switches-in between the 1st and 2nd stage of any standard guitar amp, it is easier to do than to think.
-
Even though this thread is a bit old I'd like to respond/ask about something.
The 1nFd trims the highs. This is not just clip-then-filter, the cap AT the diodes reduces clipping of high harmonics, which would throw hash up the spectrum which would be hard to filter.
In my use w/ simple diode clipping circuits like this, I've noticed and got a distinct drop off of high end w/out any "low pass" cap. The reason being the waveform's upper-most and lower-most parts are being literally clipped off from the signal. I get a nice juicy overdriven pedal-like distortion from it that helps tame the high end at low volumes when the early stage is fully bypassed. Has anyone else experienced what I have regarding this?
-
Hi jojokeo, I remembered you asking about the cap across the clipping diodes http://www.el34world.com/Forum/index.php?topic=11102.0 (http://www.el34world.com/Forum/index.php?topic=11102.0) The cap is still there, so if i get a chance on the weekend i'll will disconnect it and see if it makes a difference.
-
Thanks Timbo
-
Hi jojokeo, Disconnected the .001 that was across the diodes and can't say it did much other than maybe smooth out the OD a bit and cut out some of the noise (white noise), so i left it there. Thanks
-
Is this right that it was slightly smoother and w/ less noise without it?
-
Less noise and smoother with cap. Thanks