esaj Posted June 28, 2017 Author Share Posted June 28, 2017 (edited) 16 hours ago, Keith said: Try cutting hundreds of teeth out of plywood with a scroll saw. I got bored over Christmas and went in the completely opposite direction to @esaj - built a clock that is straight out of the 17th century. It took me nearly two months to get it working correctly. BTW, the drive weight is 13lb (6.5Kg) and it drops, by complete accident 1 inch per hour so runs for 57 hours. I made everything except the bell and the carbon pendulum rod, old bent model helicopter mainshafts came in handy for the arbors. Wow, that's really cool! I probably couldn't even cut the cogs correctly, even less make it work Here's another quite an opposite project, shelf in 6 hours: I'm not much of an artisan, to me it's function before form Mostly I just take rough measures, take a look at what I have, as usual, the shelves and the rack are made from scrap & leftovers, no materials are bought specifically, for the earlier shelf I just happened to get a 1m piece of rack rail from a friend, and make it up as I go Here I just wanted the hot air station up and out of the way (it probably needs something on the back to prevent the curtains from catching fire, though ), as well as space for other crap, and enough space deep in the back to fit an old 22" screen there, as I have an USB-microscope on the way plus it helps that I can then see the KiCad-drawing when laying out components... earlier I was jumping back and forth between the soldering desk and my computer to check what goes where (as the milled boards have no silk screen or such where the values etc. would be visible). There's just enough space between the corner-table on the right and the soldering desk to fit a desktop-computer. Still gotta do a couple more shelves for the component cabinets, the only place I can fit them is on the CNC-table (it's just temporarily on the left, goes back to its corner later), but I need the cabinets a little higher so there's space for the drawers to open... this reordering project became much larger than I initially thought Edited June 28, 2017 by esaj 2 Quote Link to comment Share on other sites More sharing options...
Hunka Hunka Burning Love Posted June 28, 2017 Share Posted June 28, 2017 (edited) So how's the Auyouie (hope that's not the sound you make when it shocks you Owwww yoouuuuuueeeeey!!) working out? Mine seems to be working well, and I like how quickly it heats up to temperature. I had a chance to try the hot air gun, and I was able to remove a SMD IC pretty easier. It was not so easy trying to reattach it as the air blew it around a lot even on the low setting. I'll have to practice that more. Edited June 28, 2017 by Hunka Hunka Burning Love 1 Quote Link to comment Share on other sites More sharing options...
Popular Post esaj Posted July 2, 2017 Author Popular Post Share Posted July 2, 2017 (edited) On 6/29/2017 at 1:04 AM, Hunka Hunka Burning Love said: So how's the Auyouie (hope that's not the sound you make when it shocks you Owwww yoouuuuuueeeeey!!) working out? Mine seems to be working well, and I like how quickly it heats up to temperature. I had a chance to try the hot air gun, and I was able to remove a SMD IC pretty easier. It was not so easy trying to reattach it as the air blew it around a lot even on the low setting. I'll have to practice that more. Somehow I had missed this post earlier... anyway, I haven't done much work with the hot air yet, just tried it on a few occasions. The BGA-nozzle that came with it (along with different sized round nozzles) wasn't good for non-BGA chips (I tried removing some ATMega328P's from junk boards), it heats up the chip too much before the solder around the legs starts to melt, I'll have to try again with round nozzles (or get a proper nozzles for non-BGA chips). I might have fried the chips (I gave up once I noticed how hot they get, I was using 330C hot air), but those weren't that important really Other than that, I've just taken apart some stuff (not chips) with the hot-air and used it for heatshrink tubes, in general it seems pretty good and the station hasn't caught fire so far . Need to practice using it more also... The soldering iron seems good too, although I think I might remove the fume-extractor, while it's otherwise nice, it's in the way when wiping the tip to a sponge. I use sponges, although I have those copper/brass/whatever cleaning balls too, I feel that the sponges are better at cleaning the tip. I've been riding, drinking beer & barbecuing mostly for the last few days (I'm on vacation), but have managed to get stuff done too: This "side" of the room is pretty much done... Last night I soldered some wiring for the lighting on the soldering desk. Did pretty piss poor job at it, as I was drunk enough not to really see straight, but luckily it doesn't carry that much current and at least shouldn't short-circuit or anything There's still stuff that's looking for a place (and random crap I've been working with, that I need to clean up ), and I need some weighs on top of the left-side shelf so I can place the component boxes on their sides (easier to pull whatever's needed out from there without "shuffling" through the pile), seen at the left edge of the picture. One drawer with wheels is sitting in the middle of the room currently, but I'll either tuck it next to table on the left or push it under the CNC-table. The combo hot-air/soldering station sits nicely above the soldering space, as well as the sample books holding the common SMD-components I use, and the desk has a lot of light to help see things (led tapes under the shelf and the desk-lamp with adjustable 10W leds and swiveling arms, shown earlier). There's an extra PC (my old work station, HP Z400) sitting between the soldering- and cnc-tables, it's still missing an OS, thought I'd drop in some *Ubuntu on it (like Lubuntu or whatever), as I mostly need just to be able to browse datasheets, view Kicad schematics and run LTSpice (which should work just fine with Wine). It'll be a lot more handier not having to jump between my main workstation and the soldering desk. Haven't received the USB-microscope yet, that's meant to be used with the PC too for the small stuff and inspecting boards. The only place where I found space to fit the component cabinets was on the CNC-table. I originally thought of putting them on the top-shelf of the measurement table on the left (in earlier picture), but the top drawers would have been too high, plus I would have needed to secure them somehow (so I don't pull one down on top of myself by accident or something ). And it could only fit three cabinets. Had to build a couple more make-shift -shelves, as the lowest drawers would be otherwise impossible to open with the CNC sitting on the table. While simple, it actually took quite long (well, "long", as in about 6-7 hours) to make those, because the holes for the bolts & screws on the bottoms of the cabinets are in really awkward places... Managed to do it though, with lots of cursing. And also the CNC-table is a lot less messier now. Still missing lighting, I have to figure out how I can fit overhead lighting above the CNC-machine without it being in the way. Today I installed the extra lighting on the soldering desk, changed the CNC-motor and made a new "bit block" that can fit the entire array of bits I have (also noticed that I don't have drills between 1.6...2.0mm), so I don't have to rummage through the boxes holding the extra tips and sometimes when I need tips that are much more rarely needed: The hole-rows are not exactly straight, as I drilled it by hand, probably should have used the CNC itself... Also, the top row doesn't have enough holes to actually fit 0.3mm-3.0mm, looks like I can't count But now I've got the entire array at my hand at all times. Also changed the motor on the CNC back to the old one (the original), as it's much quieter and I really couldn't trust the high-speed motor to do good cuts at high speeds... sometimes it worked good, but if I needed to make thin traces or such, I had to drop the feed speed or risk having to do the isolation again each time (wasting time & copper clad boards, when I was supposed to be saving time). Also got a new <10k RPM higher powered motor for the CNC, but I haven't got the ER-11 precision collet for it.... The axle on the original motor jammed when I was installing the precision collet on it, but now that I tried to remove it, it actually unjammed the axle (couldn't get the collet removed though, looks like it's there for good ). Likely I'll order another collet (or is it chuck? I always mix them...) for the new motor and use the original in the meantime, or put the fast one back in if need be, it's actually pretty fast to switch motors when you have the wiring soldered into them already. I'd expect to get the last bits done on the next few days, and can then finally get back to my projects, I didn't plan this reordering & building shelves and whatnot part to take this long originally Edited July 2, 2017 by esaj 1 3 Quote Link to comment Share on other sites More sharing options...
Rehab1 Posted July 3, 2017 Share Posted July 3, 2017 You have a great selection of bits! Well organized! Did they all come in a package or did you individually pick each one for a specific purpose? 1 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 3, 2017 Author Share Posted July 3, 2017 (edited) 48 minutes ago, Rehab1 said: You have a great selection of bits! Well organized! Did they all come in a package or did you individually pick each one for a specific purpose? I've bought certain sizes I use more in "bulk" (as in, 10-piece sets of same size), as well as "assortments", that might be like 0.3-1.2mm drills (10 pieces), one of each. I mostly use 20-degree and 30-degree / 0.1mm V-bits for isolation, so I have lots of those. Some "odd"-sizes like 0.2mm or 1mm wide (I actually ordered 0.5mm that would seem most useful, but the bits never arrived ) for wider isolations or if I want to scrape off all excess copper of the board. Typical drills needed in boards are between 0.6-1.0mm, 1.1-1.3 for larger connectors and such. 3.0-3.2mm for M3-screws for attaching the boards to stands and such. End mills are used for cutting the board off the copper-clad ("edge cut") and making all kinds of non-round -holes in the board (like cutting the "slots" in the typical DC-connector). They're kinda like drills, except they don't have the "spiral" of the drill, but a more comples multi-flute patterning or just rough pointy bits that can cut sideways, a picture of a fish-tail end-mill will probably make it cleared what I mean (although larger size than what I mostly use, I typically use 1mm titanium-coated for cutting the board out of the copper-clad, and smaller ones for cutting all sorts of larger holes): The boxes seen on the right side of CNC-machine in the picture showing the machine & component drawers hold the rest of my bits, I've racked up quite an amount of those (for example I probably have something like closer to 100 pieces of 30-degree / 0.1mm bits ). Cutting harder material like FR4 (fiber glass reinforced laminate) dulls the bits fast, but they're usually something like 2-4 euros / 10 pieces. Using titanium-coated bits with hard materials helps, but they still won't last very long with FR4, the end-mills don't seem to dull as easily as drills (although the drills seem fairly long lasting, typically they break due to my own error, like moving the drill so that it hits a holder or the board surface) and V-bits (too high feeds break the tip if the motor's not spinning fast enough, especially with smaller angle, and they also get dull and make bad cuts over time). Edited July 3, 2017 by esaj 1 1 Quote Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted July 3, 2017 Share Posted July 3, 2017 41 minutes ago, esaj said: I've bought certain sizes I use more in "bulk" (as in, 10-piece sets of same size), as well as "assortments", that might be like 0.3-1.2mm drills (10 pieces), one of each. I mostly use 20-degree and 30-degree / 0.1mm V-bits for isolation, so I have lots of those. Some "odd"-sizes like 0.2mm or 1mm wide (I actually ordered 0.5mm that would seem most useful, but the bits never arrived ) for wider isolations or if I want to scrape off all excess copper of the board. Typical drills needed in boards are between 0.6-1.0mm, 1.1-1.3 for larger connectors and such. 3.0-3.2mm for M3-screws for attaching the boards to stands and such. End mills are used for cutting the board off the copper-clad ("edge cut") and making all kinds of non-round -holes in the board (like cutting the "slots" in the typical DC-connector). They're kinda like drills, except they don't have the "spiral" of the drill, but a more comples multi-flute patterning or just rough pointy bits that can cut sideways, a picture of a fish-tail end-mill will probably make it cleared what I mean (although larger size than what I mostly use, I typically use 1mm titanium-coated for cutting the board out of the copper-clad, and smaller ones for cutting all sorts of larger holes): The boxes seen on the right side of CNC-machine in the picture showing the machine & component drawers hold the rest of my bits, I've racked up quite an amount of those (for example I probably have something like closer to 100 pieces of 30-degree / 0.1mm bits ). Cutting harder material like FR4 (fiber glass reinforced laminate) dulls the bits fast, but they're usually something like 2-4 euros / 10 pieces. Using titanium-coated bits with hard materials helps, but they still won't last very long with FR4, the end-mills don't seem to dull as easily as drills (although the drills seem fairly long lasting, typically they break due to my own error, like moving the drill so that it hits a holder or the board surface) and V-bits (too high feeds break the tip if the motor's not spinning fast enough, especially with smaller angle, and they also get dull and make bad cuts over time). You intrigue me. You are a programmer. Do you program at home? Also you have more electronic supplies than any hardware engineer I know. How many things do you build? I am jealous. 2 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 3, 2017 Author Share Posted July 3, 2017 (edited) 7 hours ago, Carlos E Rodriguez said: You intrigue me. You are a programmer. Do you program at home? Well, I work mostly from home, so yes, technically I program at home But you probably meant if I do it in my own time, I used to write lots of software just for fun, but not so much in the past several years. I "get" to program enough for work, although I like embedded programming ("fiddling bits" ) more than your typical server/desktop software and such, but still usually concentrate more on the electronics and just make a simple & straightforward software to perform the necessary logic, using available libraries rather than trying to write everything from scratch myself (there are exceptions, if I really need to get more speed out from an Arduino/ATMega or sometimes just want to try my hand at writing things from scratch myself). Quote Also you have more electronic supplies than any hardware engineer I know. This might be influenced that people working on electronics might not be as interested doing it at home on their own time, or even when they do, they get access to high quality equipment at work, and don't see the need to buy their own, plus really high-end professional equipment is very expensive, prices go fast to 5 or even 6 figures for something like modern wideband spectrum analyzers capable of tens of gigahertz and such . My hardware is mostly either cheaper hobbyist-grade stuff (like the CNC and the hand-held multimeters) or used industrial stuff (PSU, bench multimeter), that's available for cheap (at least relative to the price of a new piece) from time to time, so it's not horribly expensive. In total, I've put something between 1000-1500€ into hardware over the last one and a half years, most of the cost being racked up by the oscilloscope and the bench multimeter. There's also pieces I've made myself that I use (like the ATX-powered bench PSU and an AD9851-based function generator), but if not counting the hours spent building them, they usually have parts that are worth maybe 5-20€ each. They're not high-end, but work well enough for my needs The components I've bought have probably totaled more than the hardware, but they're bought over time, so it's not like I dropped thousands on this stuff in an instant. Quote How many things do you build? I am jealous. Umm, lots... I have a big pile of just tossed boards and other scrap, either prototypes, broken beyond (sensible) repair or not working as intended. Also I often build circuits on breadboards just for testing out stuff, but never make actual boards for those. Not sure if it's worth being jealous about, nothing I've done is really advanced, and most of it could be done just with a little knowledge of programming (when there are MCUs involved) and electronics, strip/matrix boards, a soldering iron and Aliexpress parts (or just on breadboards for temporary testing / trying out things). And like shown above, if you want to, you can build a "lab" for yourself relatively cheaply. Edited July 3, 2017 by esaj 3 Quote Link to comment Share on other sites More sharing options...
Rehab1 Posted July 3, 2017 Share Posted July 3, 2017 Do the controllers on the right of the CNC allow you to manually control the x, y and z coordinates? 1 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 3, 2017 Author Share Posted July 3, 2017 (edited) 2 hours ago, Rehab1 said: Do the controllers on the right of the CNC allow you to manually control the x, y and z coordinates? No, although it would be possible to make such a controller. The boxes there are the spindle-motor speed control with on/off -button and coarse + fine -potentiometers, the main control-board could also control the spindle speed, I've separated it so I can control it manually during the process with my own spindle-driver board, but also left the option to connect it directly to the mainboard running the g-code. The other one is just the start/continue (green), pause (blue) and stop (red) -buttons. The stop-button is important to have "at hand", rather than relying on pushing the stop in the software with mouse (really hard to hit in an emergency, like the head moving the bit towards a holder at a fast rate ) The mainboard just reads text-commands (G-code https://en.wikipedia.org/wiki/G-code ) over the USB serial link, so it would be possible to build a control board that has a joystick(s) and/or buttons or whatever and an MCU (like Arduino/ATmega) reading those and writing the commands to the mainboard via USB to manually control the x/y/z -axes or to modify the software of the mainboard itself (it's an Arduino, the software is GRBL, available here: https://github.com/grbl/grbl ) to support putting external buttons/joysticks/etc on free pins or reading from an auxiliary board over some simple protocol (so the USB-port would be left free for the PC-connection). The actual control of the process is usually done with a separate software reading the toolpath-data running on a computer and sending it to the control board, there are multiple available, I've used bCNC myself: The process for making a board is pretty much Design the board with some EDA-software ("Electronic Design Automation", like Kicad, Pads or Eagle) http://kicad-pcb.org/ Export the Gerber- and drilling-files from the EDA-software Import the Gerbers and plan the toolpaths with some CAM-software ("Computer Aided Manufacturing"), I use FlatCAM http://flatcam.org/ Export g-code files for the CNC control software Run the milling job with the CNC control software, I use bCNC https://github.com/vlachoudis/bCNC Of course there are more finer details, especially when it comes to the CAM and milling process (calculating milling depths for the V-bit angles and widths, single- or multiple passes, possibly both in horizontal and vertical movement, exporting the files, probing + autoleveling, tool changes...). I've been asked for some tips in private messages a couple of times, maybe I should some day try to write a small tutorial of how I do it, although my way probably isn't the best, but has been "refined" by lots of trial and error Edited July 3, 2017 by esaj 3 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 4, 2017 Author Share Posted July 4, 2017 (edited) To see that the old motor indeed does actually work, I did something simple: It's a holder for my tweezers, basically just a block of wood with slots milled onto it to hold the ends of the tweezer handles. Still used that old minilaptop to control the CNC, as I haven't installed any OS on the desktop yet... Also shot a video of milling it, but I sped it up for the most parts, since it's mostly pretty boring The motor's noisier than before, so it did take some damage when I installed (hammered ) the precision chuck onto the axle. The entire axle was jammed before, but trying to remove the chuck, it got unjammed and at least it still runs, although maybe with more friction than before. Edited July 4, 2017 by esaj 2 Quote Link to comment Share on other sites More sharing options...
Hunka Hunka Burning Love Posted July 6, 2017 Share Posted July 6, 2017 (edited) To reduce bit wear I wonder if spraying a little oil at the cutting end of the drill bit might help in making them last longer. You'd need a catch tray and some splatter protection though. Don't know if it would help that much. Maybe some sort of drip system that oils the bit as it cuts might work? BTW where did you end up finding replacement motors? Edited July 6, 2017 by Hunka Hunka Burning Love 2 Quote Link to comment Share on other sites More sharing options...
The Fat Unicyclist Posted July 6, 2017 Share Posted July 6, 2017 2 hours ago, Hunka Hunka Burning Love said: To reduce bit wear I wonder if spraying a little oil at the cutting end of the drill bit might help in making them last longer. You'd need a catch tray and some splatter protection though. Don't know if it would help that much. Maybe some sort of drip system that oils the bit as it cuts might work? No. use a proper cutting fluid... https://en.wikipedia.org/wiki/Cutting_fluid It makes such a difference! 3 Quote Link to comment Share on other sites More sharing options...
Rehab1 Posted July 6, 2017 Share Posted July 6, 2017 1 hour ago, The Fat Unicyclist said: No. use a proper cutting fluid... https://en.wikipedia.org/wiki/Cutting_fluid It makes such a difference! Cutting oil is excellent! It not only lubricates the bit but also keeps it cool. I use it often especially when drilling through hard materials such as steel and stainless steel. 2 Quote Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted July 6, 2017 Share Posted July 6, 2017 (edited) On 7/4/2017 at 9:51 AM, esaj said: To see that the old motor indeed does actually work, I did something simple: It's a holder for my tweezers, basically just a block of wood with slots milled onto it to hold the ends of the tweezer handles. Still used that old minilaptop to control the CNC, as I haven't installed any OS on the desktop yet... Also shot a video of milling it, but I sped it up for the most parts, since it's mostly pretty boring The motor's noisier than before, so it did take some damage when I installed (hammered ) the precision chuck onto the axle. The entire axle was jammed before, but trying to remove the chuck, it got unjammed and at least it still runs, although maybe with more friction than before. When istalling and removing press fit parts you should use cold and heat. Apply cold to the part that need to be smaller and apply heat to the part that need to be larger. The freezer is a good place to put shafts. It will give the dimension tolerance to allow easy fitting. Never use hammering ever. Use pullers. Bearing pullers.You can buy or make some with bolts nuts and brackets. Edited July 6, 2017 by Carlos E Rodriguez 3 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 6, 2017 Author Share Posted July 6, 2017 (edited) 3 hours ago, Carlos E Rodriguez said: When istalling and removing press fit parts you should use cold and heat. Apply cold to the part that need to be smaller and apply heat to the part that need to be larger. The freezer is a good place to put shafts. It will give the dimension tolerance to allow easy fitting. Never use hammering ever. Use pullers. Bearing pullers.You can buy or make some with bolts nuts and brackets. Yeah, the chucks should be heated so they expand (I think the hole is something like a few tenths of a millimeter smaller than the axle diameter, on purpose)... I tried it, but a soldering iron couldn't make it hot enough, probably would need some sort of a torch. I installed the chucks by heating them with the iron and tapping the axle of the motor (so not like hammering really hard, but I know you shouldn't bang them anyway ). The other one went in fine that way, but the other want was harder to get in, and likely the motor bearings took some hit. Originally it was stuck (like really hard to turn the axle), so I just used the other motor until now (up to 18k RPM, very loud and had lots of runout at faster RPMs, so still had to keep the RPMs low). I ordered a new motor lately, a 9k RPM one, and was thinking of removing the chuck of the stuck motor... couldn't get it off, but the motor unjammed and seems to work just fine right now (except it's noisier than before), no runout even at max RPMs (which is something like 6-7k). Just yesterday I made a board with it and today engraved the texts on a guitar pedal aluminum casing. Luckily the motors aren't that expensive (these are cheap brushed 775's, cost something like 7-15€ per piece), the precision chuck + collet cost about as much or more... Edited July 6, 2017 by esaj Typos 2 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 6, 2017 Author Share Posted July 6, 2017 (edited) 18 hours ago, Hunka Hunka Burning Love said: To reduce bit wear I wonder if spraying a little oil at the cutting end of the drill bit might help in making them last longer. You'd need a catch tray and some splatter protection though. Don't know if it would help that much. Maybe some sort of drip system that oils the bit as it cuts might work? BTW where did you end up finding replacement motors? Simple, just look for 775 (it's a form factor) motors with suitable voltages (I use 20V laptop-charger to run the motor) in Aliexpress 16 hours ago, The Fat Unicyclist said: No. use a proper cutting fluid... https://en.wikipedia.org/wiki/Cutting_fluid It makes such a difference! 14 hours ago, Rehab1 said: Cutting oil is excellent! It not only lubricates the bit but also keeps it cool. I use it often especially when drilling through hard materials such as steel and stainless steel. As for cutting oil/fluid, the motors on this machine are nowhere near powerful enough to cut solid metals, the best I can do is engrave aluminum (like the picture in the above post) at very shallow depths (0.1-0.2mm or such). I doubt the fluid's necessary or even useful with PCBs. The typical "1oz" (1 ounce of copper per square foot) PCB has a 35µm (0.035 millimeters, or 1.37 mils) thick copper layer, the rest of the board is the laminate (FR4 = epoxy resin with fiber glass reinforcement, the cheaper stuff I use for simpler boards is bakelite), ranging from 0.6 to 1.6mm or so in thickness. It doesn't seem to be the copper that dulls/breaks the bits, I can make dozens of boards on same bit with the bakelite copper-clads and it doesn't seem to even become dull, but the FR4 eats V-bits a lot faster. Typically the problem is that the very end of the V-bit snaps off, for example on the FR4-boards, I do 0.095mm cuts in two passes (0.0475 + 0.0475mm), otherwise the end will probably snap at the middle of the job even with relatively low feed-speeds. But even in two passes the bit starts to dull after some time... The laminate seems to simply be so hard that even though the V-bits are tungsten-carbide (some with titanium coating), they won't last very long, but the drill-bits seem to take a lot more (ab)use. Not that I'm complaining, since they aren't expensive, and after finding settings that work fine with FR4, it's rarely an issue anymore... of course, I do curse a lot when the bit goes SNAP halfway into 2-hour job and I have to start editing out the already done parts from the G-code as well as checking that the header commands and commands after them make sense and it doesn't push the bit down and run to the other side of the board cutting a groove when starting over . Probably it would help also to start using lower depth, I've picked 0.095mm mostly because with 30-degree / 0.1mm bit, the top-portion width of the cut is 0.15mm (helps with design), and cutting about 3 times the thickness of the copper-layer makes it certain that the isolation is done "all the way", even if the board has some "give" at some point when actually milling, that didn't come up in auto-leveling. Drill-bits made of HSS are useless with FR4, people report HSS drill-bits wearing out in a couple of dozen holes or so. I've used my current 0.8mm & 1mm (the most common sizes in most boards) carbide drills to drill probably >1000 holes each so far, and they're still working fine . Plus I've probably got at least 8-10 spare bits each left. I probably should work (at some point, right now I've got a guitar-pedal to finish) on that PCB-milling tutorial rather than edit this post endlessly with tiny details... Edited July 6, 2017 by esaj 2 Quote Link to comment Share on other sites More sharing options...
Rehab1 Posted July 7, 2017 Share Posted July 7, 2017 57 minutes ago, esaj said: I probably should work (at some point, right now I've got a guitar-pedal to finish) on that PCB-milling tutorial rather than edit this post endlessly with tiny details... Do not reply!! Finish the pedal! The layout of the text on the pedal control panel is perfect! I'm a stickler on symmetry and esthetic placement of labeling. I like how you used the level horizontal approach on labeling the top controls and then neatly staggered the text on the bottom 3 controls. 1 Quote Link to comment Share on other sites More sharing options...
Popular Post esaj Posted July 8, 2017 Author Popular Post Share Posted July 8, 2017 (edited) On 7/7/2017 at 3:37 AM, Rehab1 said: Do not reply!! Finish the pedal! The layout of the text on the pedal control panel is perfect! I'm a stickler on symmetry and esthetic placement of labeling. I like how you used the level horizontal approach on labeling the top controls and then neatly staggered the text on the bottom 3 controls. That was a busy three days... I originally thought that I wouldn't make it (the friend who I was building the pedal for had a gig in this town last night), but on late Tuesday/Wednesday -night, I picked up the old ShredMaster -plans I made late last year and quickly whipped up corrections (Replaced the non-working tone-control, added inverting amplifier on output as originally the volume was too low, improved distortion drive). I have lost the simulations of the circuits (actually, I think they're buried somewhere, I changed computers in between and probably tucked the "old stuff" to some obscure corner of the hard drive and couldn't find them right then and there), so I had to piece it together from the other prototype-distortion board I made before and some quick back-of-the-hand calculations. With about two and a half days to go (had to be done by Friday evening), I was pushing it close... I don't have recordings of the pedal, and I don't have the pedal either anymore (handed it over last night). I finished the damn thing around 7PM last night (Friday), a good hour before we had to leave for the gig. There were a couple of things I wasn't really happy about in it (the distortion overdrives too much if the pot is turned all the way, which might be down to using TL074 instead of LM324, and the softness-setting isn't that good with LEDs, probably should have gone with silicon-diodes, there's an unnecessary 2.5V bias before the delay-section etc.). I put down some markings on the schematic after-the-fact (today), so I know to fix those if I ever make another one. The delay-section is actually from http://www.valvewizard.co.uk/smalltime.html The actual op-amp used was TL074, not LM324 like the schematic says (same pinout in both), didn't have time to try out different op-amps and how they affect the sound. LM324 might have been better, as it can go down to ground-level on the output (but isn't "real" R-to-R either), leaving more headroom on the distortion-section, and it then might not clip "too much" if driven all the way (on the actual pedal, the sound becomes somewhat choppy if the distortion is dialed close or all the way to full). Board layout. I had to move stuff around to fit some of the changes, but most of it comes from the original board last fall, otherwise I would have never made it in time (the original layout took probably 10-12 hours to make, the changes were done in about 3 hours). Also, as that one was one of the first SMD-boards I'd made, I had placed everything in the back-copper, so the board had to be mirrored in FlatCAM for milling, and I had to be really careful checking which SMD-component goes where, as it's a mirror image... In all the newer SMD-boards I've made, I've learned to draw the layout on the front-copper 3D-view of board in Kicad. Due to version discrepancies (the original layout was done in older KiCad), not all components are rendered. Making g-code toolpaths in FlatCAM Actual board after milling. Note the three holes nearer to the bottom left corner in wrong places... I made some slight changes in the board after making the first toolpaths, then just redid the isolation-toolpath, failing to notice I had moved one connector. Luckily, the raster was still correct (2.54mm / 0.1"), so I could just use hand-control to drill that missing hole, and the extra one didn't matter. After the board was done, it was already wee hours of Thursday-morning. Later on Thursday, I tried lots of different valued pots (I did have rough ideas of what to use, but still had to try things) and played around with it, also removed a resistor from the bottom (marked in the schematic), as it was causing just trouble. Luckily it didn't require more modifications. The original board cutout is designed for a small cast-aluminum encasing, but I quickly realized there's no way I'm going to stuff 7 potentiometers + stomp-switch on the face of that, so I had to use a larger encasing: Holes drilled for the pots & stomp-switch on the top, input- and output-jacks are in the sides and the DC-power jack is in the back (not seen in the picture). I was planning on painting the casing and using something like stencils to make the texts in different color, but I was running out of time, so I opted just to engrave the texts: "S/N: 001"... Like I'm ever going to make a second one Board attached on the bottom with nylon-screws and nuts. I had to put a couple of silicone pads on top of each other in the bottom so they won't hit the floor. Final values for pots marked down on a piece of paper, waiting to be soldered to the final wirings. Jacks, stomp-switch & pots in place and wired, this is already on Friday-evening. The whole thing finally screwed shut and guitar jacks + DC-jack labeled. I had maybe about 10-15 minutes to try it before having to start making food & getting ready to leave... At least it still worked at that point (minus the few flaws I mentioned before, like choppy sound if distortion drive is turned out near or to maximum), hopefully it still works when the new owner gets to try it out Edited July 8, 2017 by esaj 6 Quote Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted July 8, 2017 Share Posted July 8, 2017 4 hours ago, esaj said: That was a busy three days... I originally thought that I wouldn't make it (the friend who I was building the pedal for had a gig in this town last night), but on late Tuesday/Wednesday -night, I picked up the old ShredMaster -plans I made late last year and quickly whipped up corrections (Replaced the non-working tone-control, added inverting amplifier on output as originally the volume was too low, improved distortion drive). I have lost the simulations of the circuits (actually, I think they're buried somewhere, I changed computers in between and probably tucked the "old stuff" to some obscure corner of the hard drive and couldn't find them right then and there), so I had to piece it together from the other prototype-distortion board I made before and some quick back-of-the-hand calculations. With about two and a half days to go (had to be done by Friday evening), I was pushing it close... I don't have recordings of the pedal, and I don't have the pedal either anymore (handed it over last night). I finished the damn thing around 7PM last night (Friday), a good hour before we had to leave for the gig. There were a couple of things I wasn't really happy about in it (the distortion overdrives too much if the pot is turned all the way, which might be down to using TL074 instead of LM324, and the softness-setting isn't that good with LEDs, probably should have gone with silicon-diodes, there's an unnecessary 2.5V bias before the delay-section etc.). I put down some markings on the schematic after-the-fact (today), so I know to fix those if I ever make another one. The delay-section is actually from http://www.valvewizard.co.uk/smalltime.html The actual op-amp used was TL074, not LM324 like the schematic says (same pinout in both), didn't have time to try out different op-amps and how they affect the sound. LM324 might have been better, as it can go down to ground-level on the output (but isn't "real" R-to-R either), leaving more headroom on the distortion-section, and it then might not clip "too much" if driven all the way (on the actual pedal, the sound becomes somewhat choppy if the distortion is dialed close or all the way to full). Board layout. I had to move stuff around to fit some of the changes, but most of it comes from the original board last fall, otherwise I would have never made it in time (the original layout took probably 10-12 hours to make, the changes were done in about 3 hours). Also, as that one was one of the first SMD-boards I'd made, I had placed everything in the back-copper, so the board had to be mirrored in FlatCAM for milling, and I had to be really careful checking which SMD-component goes where, as it's a mirror image... In all the newer SMD-boards I've made, I've learned to draw the layout on the front-copper 3D-view of board in Kicad. Due to version discrepancies (the original layout was done in older KiCad), not all components are rendered. Making g-code toolpaths in FlatCAM Actual board after milling. Note the three holes nearer to the bottom left corner in wrong places... I made some slight changes in the board after making the first toolpaths, then just redid the isolation-toolpath, failing to notice I had moved one connector. Luckily, the raster was still correct (2.54mm / 0.1"), so I could just use hand-control to drill that missing hole, and the extra one didn't matter. After the board was done, it was already wee hours of Thursday-morning. Later on Thursday, I tried lots of different valued pots (I did have rough ideas of what to use, but still had to try things) and played around with it, also removed a resistor from the bottom (marked in the schematic), as it was causing just trouble. Luckily it didn't require more modifications. The original board cutout is designed for a small cast-aluminum encasing, but I quickly realized there's no way I'm going to stuff 7 potentiometers + stomp-switch on the face of that, so I had to use a larger encasing: Holes drilled for the pots & stomp-switch on the top, input- and output-jacks are in the sides and the DC-power jack is in the back (not seen in the picture). I was planning on painting the casing and using something like stencils to make the texts in different color, but I was running out of time, so I opted just to engrave the texts: "S/N: 001"... Like I'm ever going to make a second one Board attached on the bottom with nylon-screws and nuts. I had to put a couple of silicone pads on top of each other in the bottom so they won't hit the floor. Final values for pots marked down on a piece of paper, waiting to be soldered to the final wirings. Jacks, stomp-switch & pots in place and wired, this is already on Friday-evening. The whole thing finally screwed shut and guitar jacks + DC-jack labeled. I had maybe about 10-15 minutes to try it before having to start making food & getting ready to leave... At least it still worked at that point (minus the few flaws I mentioned before, like choppy sound if distortion drive is turned out near or to maximum), hopefully it still works when the new owner gets to try it out You are very skillful. 2 Quote Link to comment Share on other sites More sharing options...
Rehab1 Posted July 12, 2017 Share Posted July 12, 2017 (edited) On 7/8/2017 at 0:20 PM, esaj said: "S/N: 001"... Like I'm ever going to make a second one OMG...strong stuff buddy! I like the S/N: 001. Your friend will own an original electronic masterpiece! You need to engrave an Esaj logo on the cover. I know it was a labor of love but how much time do have invested it the pedal? Can you please video a test session. I would love to hear all of the synthesized sound variations Edited July 12, 2017 by Rehab1 1 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 12, 2017 Author Share Posted July 12, 2017 On 7/9/2017 at 0:00 AM, Carlos E Rodriguez said: You are very skillful. Thanks, but still long ways to go and lots of stuff to learn 7 hours ago, Rehab1 said: OMG...strong stuff buddy! I like the S/N: 001. Your friend will own an original electronic masterpiece! You need to engrave an Esaj logo on the cover. I know it was a labor of love but how much time do have invested it the pedal? Can you please video a test session. I would love to hear all of the synthesized sound variations On this particular piece, I'd say something like 30 hours to make it, starting from modifying the original design last Tuesday/Wednesday-night and putting the last potentiometer knobs on at the Friday-evening. But, since it's based on a couple of earlier designs (that were never encased, but made first onto breadboards and then with milled boards) and lots of trying things and measuring on breadboards and reading about the stuff, I think the total's somewhere in three-figure amounts of hours for pedal circuits alone. Unfortunately, I have no recordings of this pedal, I got it made about an hour before I had to leave, and still had to do other stuff before leaving, so I didn't have much time to try the finished pedal and didn't do any recordings of earlier testing. If I do a second one (or MK3 ), I'll try to make some recordings. 1 Quote Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted July 12, 2017 Share Posted July 12, 2017 11 hours ago, esaj said: Thanks, but still long ways to go and lots of stuff to learn On this particular piece, I'd say something like 30 hours to make it, starting from modifying the original design last Tuesday/Wednesday-night and putting the last potentiometer knobs on at the Friday-evening. But, since it's based on a couple of earlier designs (that were never encased, but made first onto breadboards and then with milled boards) and lots of trying things and measuring on breadboards and reading about the stuff, I think the total's somewhere in three-figure amounts of hours for pedal circuits alone. Unfortunately, I have no recordings of this pedal, I got it made about an hour before I had to leave, and still had to do other stuff before leaving, so I didn't have much time to try the finished pedal and didn't do any recordings of earlier testing. If I do a second one (or MK3 ), I'll try to make some recordings. Maybe you could do a rev2 with all the final fixes and maybe make 10 and put them in ebay. If this tool is practical improvement for a guitarist, i am sure it would sell well. 1 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 12, 2017 Author Share Posted July 12, 2017 15 minutes ago, Carlos E Rodriguez said: Maybe you could do a rev2 with all the final fixes and maybe make 10 and put them in ebay. If this tool is practical improvement for a guitarist, i am sure it would sell well. Not a bad idea, but I doubt worth it. Even if I could make one in, say, 20 hours, it would be 200 hours of work to do 10 pieces. The components and materials aren't that expensive, probably around 15€ total, but shipping from Finland to abroad is relatively expensive, even for the smallest packages it's between 23€ and 41€, depending where it's being sent. That's for "normal" packages, Express is more expensive, starting around 40€. So just to break even, I'd have to charge between about 35€ and 55€ for each, and more to make any profit out of it. Around that price range, you can get a new quality commercial pedal from eBay or such, cheaper for used ones. Guess which one people would buy, a proven, well-known pedal from a bigname-manufacturer like Boss or Ibanez, or a home-made pedal from some random guy? 3 Quote Link to comment Share on other sites More sharing options...
steve454 Posted July 12, 2017 Share Posted July 12, 2017 On 7/8/2017 at 11:20 AM, esaj said: "S/N: 001"... Like I'm ever going to make a second one That's great, your friend gets the first one, limited edition, number one of a limited production. It looks very good. 2 Quote Link to comment Share on other sites More sharing options...
esaj Posted July 13, 2017 Author Share Posted July 13, 2017 (edited) Tried my hand at some organ harvest... I mean component salvaging today: Some gate drivers, AtMega328's, whatever the USB-interfacing -chip in Arduino Nanos is, current sensors, some SMD-power resistors, a couple of crystals... large caps. Blew off (by accident) some smaller components (I think they're 0402's, the really tiny ones in the middle), just tossed them as I can't solder them anywhere Mostly this was just practicing using the hot-air for desoldering. Not sure if the components actually survived, although after initial trying with different nozzles, air flows and temperatures, I got pretty fast at it (10 seconds of heating before getting a *QFP-chip off), the components usually still got pretty hot. Couldn't get the large (and dead) ARM-chip off the Firewheel-board, I guess it's glued there in the pick & place -phase. Also picked up the USB-microscope today, it took its sweet time to arrive (usually registered mail arrives in about 2 weeks, this took something like 4-5 weeks). It's not very good, they claimed it's 2MP, so FullHD-resolution, when in reality it's 0.3MP (640x480). Not unexpected, as it was so cheap If there is an actual zoom, it seems to be controllable only by the software, which of course doesn't run on Linux. Didn't try the thing with its own software on Windows, since it's meant to be used by the soldering station anyway. I can somewhat adjust the zoom by moving the microscope up and down on its stand (which is a bit flimsy, but "stable enough" plastic jig), and while definitely not high-end, it's not actually that bad for the price (11.85€ / about $13.50 with shipping): Close up on some 0402's Farther away... the lights are glaring off the board & coating. About as close as I could get. The component size is about 1mm x 0.5mm (0.04 x 0.02 inches). So it looks like I can zoom it to show roughly about 2 x 3mm area, which should be plenty. This is about as far as I can get. Tried also the "calibration ruler" that came with it, a thin plastic strip with some lines and widths marked on it: Again the lights are glaring from the plastic surface. A close up on 0.076mm line Close up on the text on different surface. So, not that bad all in all, a more "proper" microscope would have cost easily 10 times more, and I think I will mostly use this to read small texts off the components and maybe when soldering some tighter pin-pitch chips and smaller than 0805's. Another thing that passed my mind was that I could make a stand of my own with a stepper-motor controlling the height (like a simple linear actuator) and make it easier to adjust (the roller used to move the thing up and down is pretty tight), but not sure if that is necessary, at least as long as the original stand doesn't break Edited July 13, 2017 by esaj 2 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.