|
Post by gckelloch on Aug 14, 2022 2:58:30 GMT -5
By the way, and for those who might get a kick out of this, the name " Ash Lab" is inspired by a character in a 1979 film. What character, in what film? Why that character? Ash is a character in the film. It turn out "he" is an android, and the android gets its head knocked off in a struggle. After the struggle, the humans wish to query the android, so they hook up its head to some electronics and it's able to converse, even though it's in bits & pieces (YouTube). The idea is that, like the character in the film, a guitar can be made functional despite having guts on a table instead of inside the guitar, and one can get "info" (i.e. results of wiring models) from it. ...and the award for outstanding guitar nerd in a guitar nerd forum goes to stevewf! Takes one to know one. I'm an Alien series film fanatic myself.
|
|
|
Post by thetragichero on Aug 17, 2022 18:50:13 GMT -5
probably the most popular item i sell is a proco rat modified with an lm308 op amp (like the originals had) and a three-way clipping toggle switch (stock 1n4148, no clipping diodes, one green and one red led). this is the first time I've tied the diode grounds directly to the dc jack instead of running a wire to the board. looks cooler this way
|
|
|
Post by Yogi B on Aug 22, 2022 5:22:41 GMT -5
Trying to leave GIF plots behind, now that I've realised that you can embed animations within SVGs. Since dimensions are arbitrary in vector images, I can increase the size of the animation without worrying (too much) about my Cloudinary bandwidth limit. Plus, there's zero cost for animating back and forth — with a GIF that requires double (well, double less one) the number of frames. This post also serves as a test, so please let me know if you experience any issues with the below animation. (For reference it's a volume sweep comparing modern & '50s wiring — with the former being recoloured every 10%, as per my previous animations.)
|
|
|
Post by unreg on Aug 22, 2022 15:15:15 GMT -5
Yogi B, your animation may be moving. It difficult for me to tell bc you use all colors of the rainbow. The cold and warm colors may be causing the lines to move; idk. Maybe, for animations with little movement, you could just alternate between 2 colors… maybe both warm or both cold. That may help us to better view the small movement. And, maybe have a difference at the start to give an idea of when the gif repeats.
|
|
|
Post by MattB on Sept 3, 2022 12:48:17 GMT -5
SOT-23 op amp with 30AWG kynar wire soldered to the feet.
|
|
|
Post by thetragichero on Sept 3, 2022 22:17:18 GMT -5
MattB oh fun fun... i bought a couple of the j201 smd boards to use in place of the now-discontinued j201 whenever i need it. i've also made dip op amps work in place of the inline op amps used so often in boss pedals... not fun but i learned a lot about how something like the metal zone doesn't benefit from swapping in expensive or esoteric op amps like some of the 'named' pedal modders will gladly charge you for
|
|
|
Post by MattB on Sept 4, 2022 17:04:13 GMT -5
The first really small soldering I did was soldering some SMT J201s to adapter boards. It was a lot easier than I thought it was going to be, which is what gave me the confidence to try more small stuff. I have some ideas for active circuits to go inside a guitar, so I've been trying to learn PCB design. Need to breadboard some stuff with this one and see if it works before I actually spend money on getting boards made.
It looks like through hole J201s are still being made by Interfet- Mouser has some in stock but they're over three pounds each. I just got some 2SK208-R, which I think should be a good substitute in a gain stage. (For a buffer, my new favourite JFET is the 2SK3557-6). The three 2SK208 I've tested so far all had Vgs(off) between .59 and .6V and IDSS between 0.45 and 0.5mA. I have an overdrive nearly finished that has a JFET stage- it was designed to use a J201, but I put a socket on the board so I can try a 2SK208 and see how different it is.
|
|
|
Post by Yogi B on Sept 4, 2022 23:06:39 GMT -5
The first really small soldering I did was soldering some SMT J201s to adapter boards. That reminds me, does anyone know of any other source of SMT pin strips that fit in the usual machine pin sockets? (Standard header pins are too thick.) Mouser has these: PRECI-DIP 350-PP-164-00-206101, but they're pricy for what they are. I do like that Toshiba seem to consistently offer at least one more V GS(off) / I DSS bracket per transistor type than comparable JFETs from other manufacturers. Also I think they're the only ones I've seen that include plot showing the typical relationship between the two — not that it's necessarily that accurate, but it's got to be better than interpolating the min/max specs. (I was recently looking for 2n5953 replacements, and was disappointed not to find anything from Toshiba that seemed to fit the bill.)
|
|
|
Post by MattB on Sept 5, 2022 16:29:21 GMT -5
Well I got my overdrive finished and boxed up, got everything working except the treble control, and then this happened: These pots were less than 10p each, so I can't really complain. It did at least sound good for the few minutes I had it mostly working. Should be an easy enough fix if I can figure out how I messed up the tone section.
I do like that Toshiba seem to consistently offer at least one more V GS(off) / I DSS bracket per transistor type than comparable JFETs from other manufacturers. Also I think they're the only ones I've seen that include plot showing the typical relationship between the two — not that it's necessarily that accurate, but it's got to be better than interpolating the min/max specs. I've spent a lot of of time recently trawling through various websites trying to decide what parts I should order. I found lots of JFETs that might have worked fine for what I wanted, but with a range of values big enough that they might have been useless, so it was nice to find a few that I could be reasonably confident would do what I want. From various things I've read on the web I get the impression that in practice manufacturers are capable of better tolerances than the datasheets show, but I haven't had much luck in finding real world typical values.
|
|
|
Post by MattB on Sept 6, 2022 12:33:12 GMT -5
|
|
|
Post by Yogi B on Sept 8, 2022 21:39:00 GMT -5
Yogi B, your animation may be moving. It difficult for me to tell bc you use all colors of the rainbow. The cold and warm colors may be causing the lines to move; idk. Maybe, for animations with little movement, you could just alternate between 2 colors… maybe both warm or both cold. That may help us to better view the small movement. And, maybe have a difference at the start to give an idea of when the gif repeats. The movement should be pretty obvious, maybe not perfectly clear, but you should be able to see it. I agree that some of the colours don't contrast that well with the grey, previously I've only had a single grey trace — with eleven of both, all moving, it gets a little chaotic. Below is a duplicate, now with two static colours: green for '50s, red for modern. I've also added a readout of the current percentage of the volume control to the top right — this is implemented in a fairly silly way (CSS animations can only animate attributes/properties, not content). Previous versions had issues with flickering numbers (especially in Firefox), but that seems to be fixed in the current version (at least on my machine). Adding the percentage also made it more obvious that my animation-timing-function's were off. I had to lookup to actual web standard to find out this always represents the easing function used between keyframes, rather than timing between them. Elsewhere I looked this was poorly explained implicitly describing it in terms of an animation with only two keyframes (at the end points), and the misleading name certainly didn't help. The previous animation was going: ...3%, 2%, 1%, 1%, 2%, 3%, ..., 98%, 99%, 100%, 100%, 99%, 98%... — the one below doesn't double up the extremes & goes all the way to zero (though the latter is difficult to appreciate with the current plot). Finally, if you're the nosey inquisitive type, you'll see that I've added a CSS variable, --t, set in its own <style> tag (at the very end) to easily facilitate control of the overall duration.
|
|
|
Post by unreg on Sept 9, 2022 0:38:00 GMT -5
The movement should be pretty obvious, maybe not perfectly clear, but you should be able to see it. I agree that some of the colours don't contrast that well with the grey, previously I've only had a single grey trace — with eleven of both, all moving, it gets a little chaotic. Below is a duplicate, now with two static colours: green for '50s, red for modern. I've also added a readout of the current percentage of the volume control to the top right Sry Sir Yogi B, I can’t see definite movement in the rainbow colored one AND there isn’t movement in your recent one… it’s just solid red and Volume is always at 100%. Tried tapping your second animation and that tapping of mine was pointless. I’m using Firefox on iOS. I’m walking into a building with a roof that’s about to collapse; at least, that’s what it feels like being honest here with you genius sir.
|
|
|
Post by thetragichero on Sept 9, 2022 8:39:23 GMT -5
looks fine here on brave (chromium-based) on linux and android. sounds like an apple phone issue
|
|
|
Post by newey on Sept 9, 2022 13:49:09 GMT -5
sounds like an apple phone issue Like unreg, I also have no animation or color change, all is red and stationary. I'm using Firefox on Windows 10 on my PC. However, in other threads where Yogi B has used the animation of his plots, I have been able to see it.
|
|
|
Post by unreg on Sept 9, 2022 13:51:21 GMT -5
looks fine here on brave (chromium-based) on linux and android. sounds like an apple phone issue Cool! Glad it works for you! 😀 Maybe it is? It’s an iPhone 13… not the latest… 14 came out 2 days ago?
|
|
|
Post by unreg on Sept 9, 2022 13:55:00 GMT -5
sounds like an apple phone issue Like unreg, I also have no animation or color change, all is red and stationary. I'm using Firefox on Windows 10 on my PC. However, in other threads where Yogi B has used the animation of his plots, I have been able to see it. Yes, his other animations, in other threads, have been visually animated for me too.
|
|
|
Post by MattB on Sept 9, 2022 15:02:35 GMT -5
Like unreg, I also have no animation or color change, all is red and stationary. I'm using Firefox on Windows 10 on my PC. However, in other threads where Yogi B has used the animation of his plots, I have been able to see it. I'm also using Firefox and Windows 10. Works fine for me.
|
|
|
Post by reTrEaD on Sept 9, 2022 17:11:29 GMT -5
Finally, if you're the nosey inquisitive type, you'll see that I've added a CSS variable, --t, set in its own <style> tag (at the very end) to easily facilitate control of the overall duration. I'm both nosy and inquisitive but not seeing the variable.
|
|
|
Post by unreg on Sept 9, 2022 23:14:45 GMT -5
Finally, if you're the nosey inquisitive type, you'll see that I've added a CSS variable, --t, set in its own <style> tag (at the very end) to easily facilitate control of the overall duration. I'm both nosy and inquisitive but not seeing the variable. I bet proboards cut the extra <style> tag off… bc html is probably frowned upon here since it could be used maliciously. html being the < > tags; here [ ] is probably encouraged.
|
|
|
Post by Yogi B on Sept 13, 2022 2:10:23 GMT -5
Finally, if you're the nosey inquisitive type, you'll see that I've added a CSS variable, --t, set in its own <style> tag (at the very end) to easily facilitate control of the overall duration. I'm both nosy and inquisitive but not seeing the variable. It's in the SVG file, you'll have to open it in a new tab then use the dev tools to inspect it. By default your browser will likely download the image rather than open it directly in a new tab, so you'll need to download it then (re-)open it with the browser. Dragging & dropping the file (onto a "new tab" page, or onto the address bar) is probably the easiest way to do this. The full tag is: <style>:root{--t: 10s;}</style> I'm using Firefox on Windows 10 on my PC. I'm also using Firefox and Windows 10. Works fine for me. And this is why I don't like web dev — particularly CSS issues, 'cause there's no single place that shows all encountered errors. newey, is your Firefox up to date? It’s an iPhone 13… not the latest… 14 came out 2 days ago? Ah, I think I know what the major cause the iOS issue is: support for path() in d as a CSS property. That also shows a lack of support for Firefox below version 97 (released Feb 9 th this year), and overall global support is only at 76.67% which is lower than I'd like. (Linter support for in-/valid CSS values is relatively lacking, most checks are just for property names — that's why I didn't spot this earlier.) I guess that means I'll have to work out if I can translate the CSS based animation to 'proper' SVG animation (SMIL). Like unreg, I also have no animation or color change, all is red and stationary. Do either you or unreg see the "Volume: 100%" percentage change?
|
|
|
Post by newey on Sept 13, 2022 5:44:35 GMT -5
newey, is your Firefox up to date? That could be the issue. I'm on my work laptop, auto updates are disabled by the sys admin. That's because I'm supposed to always be using the cloud-based version, which is probably up to date. I'll check it later when I get to work and sign into the cloud. The "Vol 100%" scrolls at a very high rate of speed, from 0% to 100% in like 2 seconds.
|
|
|
Post by newey on Sept 13, 2022 9:46:41 GMT -5
I was wrong on that, it actually goes from 0% to 100% and then back down to 0% and takes about 5-6 scoends total. I checked my cloud-based version of Firefox and it, too, has auto updates disabled by the admin. Version appears to be 91.13.0esr. I believe the "esr" stands for "extended support release" Not sure if that answers your question or not, Yogi B.
|
|
|
Post by unreg on Sept 13, 2022 10:02:50 GMT -5
Do either you or unreg see the "Volume: 100%" percentage change? it’s just solid red and Volume is always at 100%No sir, on my iOS that number is also stationary.
|
|
|
Post by Yogi B on Sept 13, 2022 17:27:37 GMT -5
I checked my cloud-based version of Firefox and it, too, has auto updates disabled by the admin. Version appears to be 91.13.0esr. I believe the "esr" stands for "extended support release" Until now I didn't know that Firefox had such a thing. It'd be useful if caniuse marked things like that, you can tell there's something special about version 91, because of the uptick in usage: 90 is 0.01%, 91 is 0.08%, and 92 is 0% — but you have to be looking closely to spot that. I guess that means I'll have to work out if I can translate the CSS based animation to 'proper' SVG animation (SMIL). One of the issues with SMIL is it can't have the animation loop back and forth, at least not without almost doubling the filesize.
|
|
|
Post by newey on Sept 13, 2022 18:16:36 GMT -5
Yogi B- In your graph up above at reply #362, I am seeing the colors change but the lines are static, they don't move for any of the color changes.
|
|
|
Post by unreg on Sept 14, 2022 1:31:12 GMT -5
Using version: Firefox 104.2 (17705). That’s just me; thought it may help you though; I do realize you are attempting this animation to work for a good portion of all.
|
|
|
Post by Yogi B on Sept 14, 2022 2:23:04 GMT -5
Using version: Firefox 104.2 (17705). I should have been more specific, I was referring to Firefox on Windows/Android — AFIAK all iOS browsers are built around the same WebKit browser engine, which is tied to the iOS version (that's what caniuse lists iOS versions, rather than specific iOS browser versions).
Below is a version that uses SMIL aminations which should be much more compatible. Cycling in one direction is definitely inferior to reciprocating, as it's difficult to comprehend the extremes — almost as soon as it skips back to 100% it's moving. I could slow down the overall animation, but any slower and it'll look choppier (it's currently at 10fps, which is about the lower limit of what the human eye perceives as motion) plus I feel like a complete cycle would drag on too long.
I have thought of a way I could do reciprocal motion with SMIL without doubling up the plot data (though it will still require an increased file size, but by how much I don't yet know), however it could potentially be quite flickery.
|
|
|
Post by newey on Sept 14, 2022 5:09:26 GMT -5
OK, things are moving now. I have red and green and as V descends to zero, so does the graph
|
|
|
Post by unreg on Sept 14, 2022 15:27:25 GMT -5
Congrats! Mine works now like newey described, but when volume reaches 0% the bars are around the midpoint between -18db and -24db. (The volume text is blank after that until the animation restarts with Volume at 100%.) So either the volume text is decreasing a bit, like twice as fast, too rapidly, for me, or something else is happening to cause the green and red bars to keep lowering.
|
|
|
Post by Yogi B on Sept 14, 2022 16:31:09 GMT -5
So either the volume text is decreasing a bit, like twice as fast, too rapidly, for me, or something else is happening It's probably exactly twice as fast, I was using percentage units to position the numbers which can sometimes have issues when used as vertical lengths. This seems such a case where it's calculating based on the width instead, since the image is exactly twice as wide as it is tall — that would make the numbers scroll twice as fast. I've updated my previous post, it should be fixed now.
|
|