Nice walk through a distortion circuit here...
https://prentisseffects.blogspot.co.nz/2018/01/thermionic-distortion-friedman-be-od.html (https://prentisseffects.blogspot.co.nz/2018/01/thermionic-distortion-friedman-be-od.html)
one little mistake in description seems to be the LED clipping at end of circuit is not to ground as said but too Vref.
But I'm by no means fully enlightened to be making big statements. Good read and analysis anyway i think.
Quote from: cooder on May 15, 2018, 04:56:02 PM
Nice walk through a distortion circuit here...
https://prentisseffects.blogspot.co.nz/2018/01/thermionic-distortion-friedman-be-od.html (https://prentisseffects.blogspot.co.nz/2018/01/thermionic-distortion-friedman-be-od.html)
one little mistake in description seems to be the LED clipping at end of circuit is not to ground as said but too Vref.
But I'm by no means fully enlightened to be making big statements. Good read and analysis anyway i think.
Thanks for the compliment cooder!
Also, thanks for pointing out that error. I'll check that out today. Feel free to let me know about any other issues if you see any.
As I have zero engineering background, this was very helpful! I've been building these things long enough, should probably start learning about I'm soldering
Thanks cooder for posting this, and thanks sonnyboy27 for the site! Nicely done and was a good read.
Quote from: junkemail86 on May 16, 2018, 05:21:10 AM
As I have zero engineering background, this was very helpful! I've been building these things long enough, should probably start learning about I'm soldering
I'm glad it was helpful. I studied computer engineering in college and the only way I got through my circuits courses was to relate it to guitar pedals. Prior to making that real world comparison I was absolutely lost.
I too enjoyed the read and walk through and certainly wouldn't mind reading others.
nice pedal, i am in the process of building one myself for a 1590B enclosure.
Quote from: sonnyboy27 on May 16, 2018, 04:46:58 AM
Quote from: cooder on May 15, 2018, 04:56:02 PM
Nice walk through a distortion circuit here...
https://prentisseffects.blogspot.co.nz/2018/01/thermionic-distortion-friedman-be-od.html (https://prentisseffects.blogspot.co.nz/2018/01/thermionic-distortion-friedman-be-od.html)
one little mistake in description seems to be the LED clipping at end of circuit is not to ground as said but too Vref.
But I'm by no means fully enlightened to be making big statements. Good read and analysis anyway i think.
Thanks for the compliment cooder!
Also, thanks for pointing out that error. I'll check that out today. Feel free to let me know about any other issues if you see any.
Why thanks you for doing this in the first place very good and enlightening.
Hope to see more of that in the future!
Don't understand this part about the power supply:
"C20 is smoothing the majority of any DC ripple that existed in the power supply. It's followed by three 100nF caps which are used to filter the power between pins 8 and 4 of each of the ICs to prevent any noise in the signal path."
Surely running these caps in parallel just gives 100.3uF total capacitance?
And the three connections U1, U2, U3 are all coming from the same rails.
What am I missing here?
Quote from: Philtre on May 17, 2018, 01:36:33 AM
Don't understand this part about the power supply:
"C20 is smoothing the majority of any DC ripple that existed in the power supply. It's followed by three 100nF caps which are used to filter the power between pins 8 and 4 of each of the ICs to prevent any noise in the signal path."
Surely running these caps in parallel just gives 100.3uF total capacitance?
And the three connections U1, U2, U3 are all coming from the same rails.
What am I missing here?
I'd say it's a question of layout maybe.
Placing those .1 caps close to the chips for more efficiency maybe? So 3 different for the 3 different opamps.
It's like the cistern on top of your toilet. There to provide extra water/electricity on demand, if the pipes cannot provide enough water/electricity in a hurry.
Sure, but there are 4 caps in parallel - 100uF and 3 100nF. Thus a total of 100.3uF and where these are physically placed will make no difference to this. Yet the text talks about them as if they are separate caps for each of the three ICs:
"It's followed by three 100nF caps which are used to filter the power between pins 8 and 4 of each of the ICs"
Wouldn't a 100uF and a 300nF cap be the same?
In a perfect world and on paper, yes, but a PCB layout has its own idiosyncrasies and high gains don't help matters.
A PCB trace has resistance for instance.
Quote from: Philtre on May 17, 2018, 01:46:20 AM
Sure, but there are 4 caps in parallel - 100uF and 3 100nF. Thus a total of 100.3uF and where these are physically placed will make no difference to this.
Placement actually can make a difference in some situations. And the smaller caps in parallel do add up as you say, but from what I've read in texts about PCB design, the smaller value caps can react "faster" to smaller changes in current than the larger caps. So if you have large and small caps in parallel, overall ripple noise smoothing is improved. From what I gather, it's sort of like improving the "slew rate" for power supply current transients. But I might have misunderstood some of the theory.
This kind of thing is more important with digital processing and things like power amplifiers (chip amps, etc) than it is with low-voltage / low-current analog stuff like dirt boxes. But it's just good practice to do it.
Quote from: culturejam on May 17, 2018, 07:49:59 PM
[And the smaller caps in parallel do add up as you say, but from what I've read in texts about PCB design, the smaller value caps can react "faster" to smaller changes in current than the larger caps. So if you have large and small caps in parallel, overall ripple noise smoothing is improved.
Ive seen this in the "huminator" power filter, its a common tactic for power filtering (large electro + small film caps). Not sure why there are 3 x 100nf caps though, I guess more "pipes" to filter the current, faster reaction maybe, but I'm just guessing here.
Quote from: chromesphere on May 17, 2018, 11:30:25 PM
Ive seen this in the "huminator" power filter, its a common tactic for power filtering (large electro + small film caps). Not sure why there are 3 x 100nf caps though, I guess more "pipes" to filter the current, faster reaction maybe, but I'm just guessing here.
Yeah, even Brian has a 100n parallel with the larger filter cap in all his designs these days. But I agree that I'm not sure 3 of them are necessary, although I suppose if you place those very near the op amp's VCC pin, it might help reduce noise in some situations.
Quote from: culturejam on May 17, 2018, 07:49:59 PM
Quote from: Philtre on May 17, 2018, 01:46:20 AM
Sure, but there are 4 caps in parallel - 100uF and 3 100nF. Thus a total of 100.3uF and where these are physically placed will make no difference to this.
Placement actually can make a difference in some situations. And the smaller caps in parallel do add up as you say, but from what I've read in texts about PCB design, the smaller value caps can react "faster" to smaller changes in current than the larger caps. So if you have large and small caps in parallel, overall ripple noise smoothing is improved. From what I gather, it's sort of like improving the "slew rate" for power supply current transients. But I might have misunderstood some of the theory.
This kind of thing is more important with digital processing and things like power amplifiers (chip amps, etc) than it is with low-voltage / low-current analog stuff like dirt boxes. But it's just good practice to do it.
From what I remember in my board design course, it's basically what CJ said. It's a common practice to place a filtering cap between power and ground near each chip. In the original layout those 100nF caps are next to each IC. This is done to reduce any residual noise that could be introduced between the main filter cap(s) and reaching the chip itself. It's common practice in digital circuitry and so it crosses over to analog designs sometimes (most notably in higher gain circuits). It's one of those board layout things that shouldn't be an issue from a theoretical standpoint but it helps in reality.
This is good timing. You can see the effect of different bypass caps on Daves oscilloscope:
https://www.youtube.com/watch?v=1xicZF9glH0&t=628s
Quote from: chromesphere on May 18, 2018, 04:17:15 PM
This is good timing. You can see the effect of different bypass caps on Daves oscilloscope:
Brilliant! This video and Dave's other one on bypass caps explains thing nicely. Thanks for the heads up.
(Side note - some of the vero layouts at tagboardeffects remove the additional small-value bypass caps. Probably not a good idea then!)