|
all the hands sweep continuous 360o paths, yeah, nothing different in that regard. i'm dubious that modifying a commercial movement is gonna that easy, tho. it sounds very simple in principle but, man, those commercial clock movements are incredibly tiny and tightly-packed, id basically have to design a new movement housing + layout from scratch using the old components plus the new reduction gearing (which i dont think ive ever seen sold in any other context, theyre not those ubiquitous 0.5M nylon gears you can get for pennies), and that sounds like a bigger technical challenge than pulleys that largely work themselves out or whatever else i could go with AC synchronous motors can't run off a AA for months like a quartz escapement movement can, but they definitely sip power relative to steppers- a couple watts from the mains isn't make-or-break for me and was typical for historical synchronous motor clocks. that concern was more about how stupid it would be to move a stepper once per second for the next several years or whatever
|
#
?
Nov 17, 2018 04:06
|
|
|
|
| # ? May 28, 2024 09:44 |
|
|
I'm saying put the nylon gears on the outside. Yeah, I've pulled apart those movements, there's no way you're modifying something inside them. Or, put a pulley on the output, if you feel more comfortable with those. Stepper motors can be turned off in between steps, and with passive breaking, so I don't think they will be substantially more power hungry than AC motors, but I haven't run the numbers!
|
|
#
?
Nov 17, 2018 04:14
|
|
|
Ambrose Burnside posted:AC synchronous motors can't run off a AA for months like a quartz escapement movement can, but they definitely sip power relative to steppers- a couple watts from the mains isn't make-or-break for me and was typical for historical synchronous motor clocks. that concern was more about how stupid it would be to move a stepper once per second for the next several years or whatever Whatever your stepper degrees is, you can make your gear train compensate so that 60 pulses=1 full revolution. If you're going one full pulse every time, then you can just not power the thing in between pulses; the permanent magnets cog pretty well and will hold the hopefully minuscule torque of your movement. Even if the thing does 1/π steps/revolution, just doing math with pulley wheel sizes will get you so that 60 steps = 1 revolution. Then you just need your one-pulse-per-second from literally anywhere and the rest is mechanical. Get a unipolar stepper (or drive whatever stepper you find in unipolar mode) and the wiring is even easier.
|
|
#
?
Nov 17, 2018 06:38
|
|
|
Steppers can go quite small. For example, The smallest, cheapest, crappiest 4096 step motor on digikey draws 0.5W when stalled If you're willing to have a wall plug, you should be fine. It'll be hard to get months/years out of a couple AAs though
|
|
#
?
Nov 17, 2018 06:56
|
|
|
I don't know how gross this is as far as hacks go, but I did a board layout with the through-hole DIP version of a part in mind and then found out I didn't have enough vertical clearance so I created a DIP/SSOP frankenpackage and swapped it in instead of redoing the layout even a little. Also the collector pads on the left were modelled as massive surface mount pads even though they'll never have anything soldered to them.
|
|
#
?
Nov 17, 2018 12:02
|
|
|
BattleMaster posted:I don't know how gross this is as far as hacks go, but I did a board layout with the through-hole DIP version of a part in mind and then found out I didn't have enough vertical clearance so I created a DIP/SSOP frankenpackage and swapped it in instead of redoing the layout even a little. I'm usually pretty pedantic, but neither of those seems particularly hacky to me. Normally you'd have a footprint for each of DIP and SSOP, put both on the schematic, and selectively populate one, but it makes the schematic uglier, and the main benefit is that it makes BOM management cleaner, provided you're using the schematic to generate your BOM. And making a part with pads that will never be soldered is totally fine. That's how I make "connectors" for flex circuits and standardized fiducials. Just make sure they don't have anything on the paste layer if the design ever gets to the point that it needs SMT assembly.
|
|
#
?
Nov 17, 2018 17:14
|
|
|
That's kinda neat looking
|
|
#
?
Nov 17, 2018 17:17
|
|
|
So I've got an amp (SMSL SA50) that requires a 24v DC power source. I've got a bunch of batteries that I can wire in series to make a ~26v DC power source. What do I need to make the two play nice? 24v regulator? Is that it? It's DC-DC so I don't need a rectifier. Do I even need a capacitor? Would it be better (more efficient) to wire them in parallel to make ~13v and use a step up transformer?
|
|
#
?
Nov 17, 2018 18:06
|
|
|
What kind of DC-DC regulator are you thinking? I assume you're doing some sort of vehicle thing, and that whole industry likes to play a little fast and loose with terminology, so better to clarify Transformers only work with AC, that won't play any part in this system. Have you checked your amp's specs and made sure that you can't over-volt them juuust a little? Usually devices can go plus or minus ten percent, so 26v would be fine, but also assuming you're using car batteries, you'd have to expect more like 28.8 charging voltage I guess
|
|
#
?
Nov 17, 2018 18:22
|
|
|
ante posted:What kind of DC-DC regulator are you thinking? Sorry I meant converter, not transformer. I would be building my own circuit in either case unless there's a cheap off the shelf solution. I'm open to suggestion with regards to parts. No charging. At least not while playing. Some homebrew lifepo4 battery packs for motorcycles. I powered the same amp with a bunch of UPS batteries in parallel and series last year for 2 straight days. It worked great, but it sat at almost exactly 24v the whole time. I think the MAX input voltage is 26v but I'd rather play it safe and regulate the amps input voltage. Whatever I make needs to be able to handle the max power requirements of the amp. The amp outputs 50wpc RMS. I do not know what the input power req are but the power supply that came with it is capable of 108 watts (6A max) Edit: This is a suitable example I believe: https://www.victronenergy.com/dc-dc-converters/orion-ip67-24-12-dc-dc-converters GnarlyCharlie4u fucked around with this message at 18:54 on Nov 17, 2018 |
|
#
?
Nov 17, 2018 18:46
|
|
|
BattleMaster posted:I don't know how gross this is as far as hacks go, but I did a board layout with the through-hole DIP version of a part in mind and then found out I didn't have enough vertical clearance so I created a DIP/SSOP frankenpackage and swapped it in instead of redoing the layout even a little. Beautiful design for a small run you have breakout options for every pin!
|
|
#
?
Nov 17, 2018 19:58
|
|
|
I bough an osmelloscope (Rigol 1054z, and I've already hacked it). Any preferred sites to teach me how to use it? I have a "how to oscilloscope" textbook from the 1940s, in a box, somewhere, and I assume it'd be good enough, but would prefer text on the internet. If not, I could always call up my old racist uncle who used to be a TV repairman and get lessons in person.
|
|
#
?
Nov 17, 2018 21:55
|
|
Bad Angus! Bad!
|
KnifeWrench posted:I'm usually pretty pedantic, but neither of those seems particularly hacky to me. Normally you'd have a footprint for each of DIP and SSOP, put both on the schematic, and selectively populate one, but it makes the schematic uglier, and the main benefit is that it makes BOM management cleaner, provided you're using the schematic to generate your BOM. Oh yeah, I see that my electrode footprints all have solder paste associated with them. This wasn't a concern for my hand assembly but I'll get rid of that just in case so I don't forget in the future. ante posted:That's kinda neat looking shovelbum posted:Beautiful design for a small run you have breakout options for every pin! Thanks, I spent way longer than I should have tweaking it down to the mil because I wanted it to be pretty. It's going to be a production run of 1 ever for a very specialized experiment so I splurged on things like 4 layers for a fairly simple circuit just so the very weak signals coming out of the collector pads could be shielded on all sides by ground fills and gold plating. I justified the gold plating with the fact that corrosion would screw it up but it never left a bone dry climate controlled lab and will be sitting in an environment of propane-based tissue equivalent gas made of noncorrosive stuff but really it looks nice. Here's a test fit I did before doing the proper assembly. The 3D printed part holds an alpha particle source, the top board has a resistor divider on it and big electrode, and the middle board is a (gunked up mangled old) gas electron multiplier foil.  Here's what the fresh GEM foil used in the real construction looks like. These things are made of Kapton with copper plating on both sides. There are some thousands of 50 micrometer diameter holes chemically etched into each square centimeter of them in a hex pattern. It looks like solid copper when you look at it, but if you hold it up to the light it's transparent. If you put a high voltage across them (450 volts in my experiment) and put them in an environment of the right kind of gas, electrons that enter the holes experience a gas multiplication effect so for every electron that enters you get many coming out. The gain in my experiment should be in the 100 to 200 range or so, but I don't know for a fact and knowing the precise gain is not important. We bought these from CERN where the inventor, Fabio Sauli, works. This is the 5x5 cm version and they make them much larger than that.  Here's the divider and top electrode board. The top electrode is kept at -1700 volts while the collector pads on the board I showed earlier are all held at 0 volts. The divider is configured so that there's 1000 volts across the region that the particles are entering, 450 volts across the GEM foil, and 250 volts across the lower region where electrons are collected. the 1000 and 250 volt figures were selected to maintain 1kV/cm electric fields in the upper drift region and lower inductive region which gives the GEM maximum transparency to electrons according to research I found, while the 450 volts is close to the highest I can get away with before arcing happens (500 volts or so) with some margin. The "upper" side of the GEM foil has a big resistor on it while I jumpered over the resistor position for the "lower" side. In the event of huge multiplication events in the GEM foil this can suppress damaging electrical arts by causing the voltage across both sides to drop. It works because during an experiment where I turned it up to -3000 volts I was getting output that indicated rapid periodic arcing but the foil was undamaged I guess because the arcing wasn't able to be continuous.  Here's the final assembly. I used Apiezon high-vacuum wax to seal up any exposed parts on the high voltage section. I taped paper around the wires attached to the GEM because I was squeamish about dripping wax anywhere near something that would be ruined by it, but later while disassembling it to inspect the GEM for damage after the arcing test I did I removed the paper and waxed up the solder tabs because it didn't scare me as much. When I took this photo I had forgot to attach the source holder, which was intended to be superglued on one side to the divider board. Before doing the final assembly I had a "hey wait a minute" moment and fixed that.  Here's another view of the assembly. Originally each of the low voltage power connector (+/- 12 volts from a NIM crate), high voltage, test signal input, and preamp output, and selector header were going to have individual ground connections, but I realized it would be best if I could keep connections to a minimum to make the feedthroughs simpler. I reduced that to 2 ground connections, one for the selector signals and one for the rest. So in the end I grounded the high voltage board via the now-unused ground pin connected on the output signal header since it was in a convenient location.  Here's the board sitting in the vacuum chamber with the hookup wires attached. The selector board that provides the signal to the 1 to 16 multiplexer is sitting in the background. It's just some chepa stripboard with a set of DIP switches, some pull-down resistors, a 6 pin header, and a connector for a 9-volt battery. I chose a battery so I wouldn't have to bring in one of my own power supplies or source another one, plus if nothing else a battery is very low noise so I don't have to worry about it being a noise source.  Here are the vacuum feedthroughs I made. Commercially-available feedthroughs are very expensive so I based mine around reducing bushings which are normally designed to connect larger pipes to smaller ones. I 3D printed wire guides that consisted of two concentric cylinders in a "mushroom" shape designed to fit right in the middle of the bushing, with a ring of tiny holes for wires to fit through. I made the pins ouf of 21 gauge solid core wire threaded into the holes in the wire guides, which were then super glued into position in the bushings. After the glue cured I filled both sides with that high vacuum wax. Obviously this setup would have issues at high temperatures but it works beautifully at room temperatures. The blue wire is high-voltage rated wire and I didn't want to mess with connectors for 1700 volts so I permanently attached the wire to that one. The rest of the internal signals use one pin rectangular header connectors that fit onto the "pins".  Here's the inside of the "pod" with the connectors.  Here's everything all sealed up with the output of a test pulse showing on the scope, showing that the preamp and all connections related to it work. I'm limited in my ability to diagnose problems once it's sealed up and filled with gas without letting the gas out and undoing all the screws which is tedious.  So what does it even do? The chamber is filled with propane-based tissue-equivalent gas at about 1/3 atmosphere, which makes the 65 mm length the alpha particles will travel (15mm in the collimator, 50mm in the sensitive area) radiologically equivalent to the 40 micrometer length they'd travel in human tissue. I'm taking measurements of ionization (and thus energy deposition) at 16 points across the final 50mm of the particle's path. The MAX336 16 to 1 multiplexer lets me connect one collector pad at a time to the Cremat CR-110 preamp, The ionization caused by an alpha particle is substantial, but still isn't a very strong signal, so the GEM foil provides a high gain to the signal before it even reaches any of my electronics. Without it, even the 1.4 volts per picocoloumb charge to voltage conversion of the preamp wouldn't be sensitive enough to detect anything. But with the gain of 100 to 200 the GEM provides, the output is well above the noise floor. I'm using an Amptek DP5 multichannel analyzer to count the pulses coming out of the detector. It takes several hours per channel of my experiment to collect enough pulses to be statistically meaningful, and I can only possibly measure one channel at a time, so the experiment will take a few days once I start the final run of it. When I plot energy per channel, I'm expecting to see a Bragg curve like this:  Except in human tissue instead of air. My plan is to relate the resutls of my experiment to lung tissue damage from inhalation of radon. The experiment looks messy in places, but it would probably shock you to see how messy scientific apparatus looks in person. I've seen experiments where the paper makes it sound so elegant but then you see it in person and it's a hacked together junk pile that just happens to give good data. edit: Oh yeah I forgot to mention that it seems to work, but I have it hooked up to a vacuum pump over the weekend to let it offgas more and I'm going to refill it on monday and do a second longer set of test runs BattleMaster fucked around with this message at 23:41 on Nov 17, 2018 |
|
#
?
Nov 17, 2018 23:27
|
|
|
Chillbro Baggins posted:I bough an osmelloscope (Rigol 1054z, and I've already hacked it). Any preferred sites to teach me how to use it? I have a "how to oscilloscope" textbook from the 1940s, in a box, somewhere, and I assume it'd be good enough, but would prefer text on the internet. All About Circuits is my go-to for electronics education on the internet https://www.allaboutcircuits.com/worksheets/basic-oscilloscope-operation/ https://www.allaboutcircuits.com/technical-articles/basic-waveform-analysis-with-an-oscilloscope/ https://www.allaboutcircuits.com/technical-articles/an-introduction-to-oscilloscope-probes/ You do kinda need some signals to probe, though, to practice setting triggers and making measurements and stuff. Have you got an Arduino? Set it up to send serial data and probe the TX pin and play around with that. Finding analog signals is a little harder but maybe play with a microphone or a piezo cell and see what happens.
|
|
#
?
Nov 17, 2018 23:43
|
|
|
My only concern is using mains frequency for timing. I'm not that sort of electronics engineer, but my understand is that the voltage and frequency do move about more than you think. The power companies do their best to keep it under control, but I wouldn't do timekeeping based off it. The electronics to convert the mains into a usable signal are not considerably smaller than the electronics to generate a much more reliable signal from a quartz crystal, and if you're not interested in designing the electronics yourself, it's far easier and cheaper to buy an RTC (realtime clock) module that'll do what you want.
|
|
#
?
Nov 18, 2018 00:57
|
|
|
Splode posted:My only concern is using mains frequency for timing. I'm not that sort of electronics engineer, but my understand is that the voltage and frequency do move about more than you think. The power companies do their best to keep it under control, but I wouldn't do timekeeping based off it. My research says that the precision of mains frequencies varies across countries and utilities; some no longer prioritize the timekeeping utility and have lowered the accuracy requirements and nightly speed up/slow down compensation expectations, but others haven't budged. My region, from what i've read, still maintains the mains frequency explicitly for timekeeping purposes that legacy equipment relies on, and accordingly can be expected to hold better accuracy than a cheap quartz-derived signal over a months-and-years timespan (although quartz is indeed better for hour-to-hour scale accuracy). It also appeals to me on a pure retro basis, in the " *brushes dust off lapel* i'm into some pretty obscure timekeeping, you probably havent heard of it" sense i think it's all moot tho, i found out that Quartex makes a specialty 24-hour mechanism that i can pick up on amazon for cheaper than the parts to build one from scratch, so i'll probably take the path of least resistance Ambrose Burnside fucked around with this message at 01:40 on Nov 19, 2018 |
|
#
?
Nov 19, 2018 01:34
|
|
|
The US IIRC still does the timekeeping correction nightly, despite random legislation that has tried to remove that requirement (for ??? reasons ???).
|
|
#
?
Nov 19, 2018 02:41
|
|
|
In my nuclear plant classes we were told that the frequency of the generators had to be matched and in phase with the grid or bad stuff happens so I would have guess it's pretty accurate.
|
|
#
?
Nov 19, 2018 02:51
|
|
|
BattleMaster posted:In my nuclear plant classes we were told that the frequency of the generators had to be matched and in phase with the grid or bad stuff happens so I would have guess it's pretty accurate. yeah thats why DC links and batteries are so nice. telsa makes a killing from providing frequency stability in australia
|
|
#
?
Nov 19, 2018 03:05
|
|
|
BattleMaster posted:In my nuclear plant classes we were told that the frequency of the generators had to be matched and in phase with the grid or bad stuff happens so I would have guess it's pretty accurate. yeah, the idea of using synchronous AC at all was given to me by an engineer coworker who was like "if you're fine with a clock you gotta plug into the wall you can sidestep a lot of the worrying about the timing", it's a pet passion of his it seems like, he uses a mains-timed alarm clock that's like 40 years old and is very happy with the accuracy so i'm not concerned with the core viability of that approach from a timekeeping perspective Queen Combat posted:The US IIRC still does the timekeeping correction nightly, despite random legislation that has tried to remove that requirement (for ??? reasons ???). that was back in 2006 and didn't succeed. has other places, like as mentioned australia gave up on that some time ago. it really does seem to be a patchwork of standards at this point tho, that timekeeping standard used to be universal in the US and other places but not any more. the arguments for letting the frequency get sloppier I've heard focus on Finding Efficiencies And Savings (which sounds suspect given that letting things get 5% less accurate will absolutely not translate into 5% greater savings or wahtever, and it also doesnt factor for replacement costs for machines and equipment that are mains-synchronized and can't tolerate greater error), as well as to adjust for the rise of common grid-tied renewable production, which sounds more legit given that renewables produce wack-rear end fluctuating electricity that needs a lot more sculpting before oyu can dump it into the grid than a conventional steady-running generator scheme or whatever may, and the standards required can (supposedly) be legit cost-prohibitive for normal homeowners and the like compared to a more relaxed expectation from producers Ambrose Burnside fucked around with this message at 03:14 on Nov 19, 2018 |
|
#
?
Nov 19, 2018 03:11
|
|
|
BattleMaster posted:In my nuclear plant classes we were told that the frequency of the generators had to be matched and in phase with the grid or bad stuff happens so I would have guess it's pretty accurate. Yeah, all plants have to sync up to the grid so they're not at risk of interference cancelling each other out (and until they sync up they dump their power output into what is basically a massive building-sized resistor which is real cool imo) but the grid as a whole could slowly drift if everyone's just using the grid itself as the reference they're sync'd to. I think America's grid is still very very accurate but I know Europe's grid has been having weird issues with frequency drift lately that may or may not be Russia intentionally loving with it...
|
|
#
?
Nov 19, 2018 03:13
|
|
|
Though tbf I've never actually tried building a mains clock, in my case the only clock I built I went way overboard on and gave it a GPS receiver specifically so it can sync to atomic time because why not, they're like $5 these days
|
|
#
?
Nov 19, 2018 03:18
|
|
|
I designed a mains-locked clock once, going to 7400-series logic. Never built it, because it sketched me the gently caress out
|
|
#
?
Nov 19, 2018 03:30
|
|
|
Actually now that I'm thinking about it I had a clock from the 80's that I got at a garage sale that was definitely using mains timing, and I used it as my main alarm clock for years and it kept time great. It was also fun to set because it had two buttons to advance the time, one that advanced it slowly (ticking forward one minute per second) and one that advanced it at the full 60Hz and was impossible to use with any accuracy 
|
|
#
?
Nov 19, 2018 03:57
|
|
|
yeah they hang/have hung around in clocks/devices that will get plugged into the wall either way. alarm clocks w mains timing got made for decades after wall clocks went quartz, and they're -still- used for timers in various appliances, especially washer/dryer timers. mine at home def has one, poked around because I was curious apparently digital-clock-using appliances even use mains timing sometimes nowadays, i found someone complaining about how their brand new automatic coffeemaker was running exactly 1/6 slower than it should have after taking it across the Atlantic during a move. im assuming skipping out on the RTC component is marginally cheaper overall but idk
|
|
#
?
Nov 19, 2018 04:14
|
|
|
I had a mains-locked clock once, for about a day and a half. My ex and I found one of those old punchcard timeclocks (if you've played Team Fortress 2, it's pretty much the same as the one used as set dressing in the interiors) for like $15 at an antiques store. We tried using it as a mantel clock because we were really into the TF2-style midcentury aesthetic at the time, but after the first night the gears noise and clicking when it tripped over ever minute got annoying, so I gutted it and replaced the innards with a quartz movement from one of those cheap wall clocks that takes a pair of AAs (I did keep all the bits for the parts bin, there were some nice terminal blocks and motor and all). Probably more accurate, and certainly more efficient. Edit: I really wish I could've gotten the wall clock from my grandmother's bedroom after she passed away, but my aunt called dibs when divvying up the estate with my mom. That thing must've been mains-timed, it sounded like a drat factory when I'd stay the night at her house on weekends/summers when I was a kid. And aesthetically, it would've fit right in onboard a Soviet submarine in the mess hall. In other news, I used the oscilloscope to actually diagnose a thing for the first time*! I bought a couple of those little solar-powered security lights from SciPlus, finally found the "on" switch (not mentioned on the packaging) and hung the as-yet-unmounted one up right under my desk lamp for a few hours, still no joy, so I opened it up and put the 'scope to it. The solar panel seems to be working, putting out 245mV to the pads where it goes into the PCB. Seems reasonable for being on its side under an old-rear end 20W fluorescent tube. The (*My $25 Harbor Freight multimeter would've done the job in half the time, but when you've just bought a $350 hammer ...) Chillbro Baggins fucked around with this message at 10:31 on Nov 19, 2018 |
|
#
?
Nov 19, 2018 04:27
|
|
|
I still use a copal hi fi timer flip clock as my alarm, and here in the US it's very accurate. Have had to touch it except in a power outage, and it has switching outlets on the back that I've got a lamp plugged into instead of the buzzer.
|
|
#
?
Nov 19, 2018 04:31
|
|
|
Chillbro Baggins posted:The 18650 cell, however, read 950mV when I put the probes to the ends of it. That battery is dead as hell, right? Like, it's completely hosed if you let it get below ~2-3V? I guess they got 'em cheap because they've been sitting in a warehouse and self-discharging forever, and well, more likelihood of the ol' bathtub curve in that case (I will be having a word with them at some point if it doesn't come back from the dead after a day of sunlight and/or burns down the things the outside ones are mounted to). Yes, lithium batteries are generally dead (as in killed) once they get down below maybe 2.8v per cell. The specific value depends on the exact cell structure, how much load it was under when discharged, etc. but definitely at 0.9v it's no good. Realistically, it will probably still work somewhat if you only trickle-charge it with a solar panel, but it's not happy. You are sure it's a lithium cell and not just something like a NiMH in an 18650 package, right? I've taken apart some of those solar lamps and they all had either NiMH or NiCd cells because you don't need any sort of charging logic with those -- just wire them right up to the solar panel with a diode and you're done. NiMH cells are 1.2v at full charge and 0.9v would be maybe a 10-15% state of charge.
|
|
#
?
Nov 19, 2018 07:16
|
|
|
Sagebrush posted:Yes, lithium batteries are generally dead (as in killed) A. Sorry, forgot to clarify since we were talking about batteries. I believe the technical term for the condition I was describing is "turbofucked". B. Opened it back up and popped the cell out (it's clearly the same molds as used for the previous version, just with the wires soldered directly to the battery instead of spring terminals), says "LC 14500 800mAh 3.7v" on the back side. And yes, I realize I vastly overestimated the physical size of it. 14500 means AA-sized, as I confirmed by opening up my wireless mouse and looking at the AAs in there. Derp.. It's gained a quarter of a volt since my last post, leaving it pinned up under the desk lamp. So still certainly drained, but maybe recoverable with enough solar-panel trickle-charging? What chemistry is nominally 3.7v? According to some light Googling, it is Li-ion after all. Welp. I'll leave it up and see what happens, I guess. Edit: If I can't fix it, I'll send it to BigClive, this is the sort of thing he loves.  (to be fair, I could go down to the local battery store and get a replacement cell and still be under the cost of buying the same light that had a better chance of not being DOA, but that's half the fun of buying surplus, innit?)
Chillbro Baggins fucked around with this message at 11:06 on Nov 19, 2018 |
|
#
?
Nov 19, 2018 10:29
|
|
|
My clocks are basically wired displays that get timecode+1PPS over cables, fed from a set of GPS disciplined OCXOs. The distribution amplifier has redundancy switching and a NTP server built in. Why? So all the clocks tick at and show the exact same (and correct) time 
|
|
#
?
Nov 19, 2018 21:16
|
|
|
My seemingly-DOA solar light has made it up to 2.1V since my last post, maybe there's hope for it yet. In other news, I apparently pushed the "hold" button (which is a latching switch) on the cheap meter, and when I turned it on it read 6.66V  Is that just a thing with the HF meter in the red rubber bumpshield (Cen-Tech P37772), or should I contact a priest and/or the store?Also, forgot to reply to this: Sagebrush posted:You do kinda need some signals to probe, though, to practice setting triggers and making measurements and stuff. Have you got an Arduino? Set it up to send serial data and probe the TX pin and play around with that. No, but I have some FRS walkie-talkies that were all the rage in the late '90s and still work, I figure poking around in one of those ought to be interesting. Obviously I won't be able to see the oscillator chip's output, but the bit before the crystal should be within my 'scope's range (now 100MHz, does that count as  ?) Or get out an extension cord and hump the thing out to my other car that's having issues with the ignition system -- i.e., it has eight pistons, runs as a V6 (four working in one bank, two going bang and two pumping air on the other side), and the exhaust smells of raw gasoline. Having replaced the coils-on-plugs, I'm pretty sure it's the engine computer, and now I can find out! Comedy option: Take apart a dead TV laying around, see what's wrong with it. The modern ones are somewhat less likely to kill you than the CRTs my uncle worked on back when 'scopes had CRTs.. quote:Finding analog signals is a little harder but maybe play with a microphone or a piezo cell and see what happens. Anybody got a source for audio files to make Lissajous whatsits? I found all the videos explaining the principle/how to set up the 'scope, but can't find an audio source to do the input end of that thing I now know how to do on the 'scope end. Chillbro Baggins fucked around with this message at 07:58 on Nov 20, 2018 |
|
#
?
Nov 20, 2018 07:55
|
|
|
Chillbro Baggins posted:Comedy option: Take apart a dead TV laying around, see what's wrong with it. The modern ones are somewhat less likely to kill you than the CRTs my uncle worked on back when 'scopes had CRTs.. 95% of the time it's either a shattered screen or bad caps in the power supply. If you carry a flashlight you can figure out which it is before plugging them in. They never, ever have remotes though. There's somebody in my building fixing TV's 'cause all of the ones I see in the trash are opened or are the 5% with other problems.
|
|
#
?
Nov 20, 2018 08:23
|
|
|
MisterOblivious posted:95% of the time it's either a shattered screen or bad caps in the power supply. If you carry a flashlight you can figure out which it is before plugging them in. I collected garbage TV's for a while but invariably managed to get the ones without obvious flaws, where the backlight would power up and everything would appear fine but the TV bits would just... not come on... Tried debugging them but couldn't find anything obviously wrong, unfortunately. At least if you take the screen part off the backlight box makes a really nice even solid white light source that's generally PWM dimmable, you can use it as an artificial sunlight window or lightbox or just a real bright light. Also on one occasion I hooked up a probe ground to a thing I thought was ground that was actually """"ground"""" and vaporized a large portion of the circuit and the wire going to my probe which was fun, so there's that
|
|
#
?
Nov 20, 2018 15:29
|
|
|
Oh yeah, pro tip for the guy with the new scope, don't use it to test mains grounded things unless you have an isolation transformer
|
|
#
?
Nov 20, 2018 17:46
|
|
|
Yeah, this old EEVBlog goes into why if you're interested: https://www.youtube.com/watch?v=xaELqAo4kkQ I had actually seen that at the time but I guess didn't quite grasp what the message was At least it only exploded a cheap micrograbber lead, though it did blow the plastic housing bit of the micrograbber into multiple pieces that flew across the room which was... exciting. e: Oh it also made a very satisfying deep drawn out FWOOM explosion sound too for whatever reason which was actually real cool
|
|
#
?
Nov 20, 2018 17:59
|
|
|
ante posted:Oh yeah, pro tip for the guy with the new scope, don't use it to test mains grounded things unless you have an isolation transformer Oh yeah, I'm only going to be touching the low-voltage side, my TV-fixing uncle has told me enough horror stories about flyback transformers and my HVAC-fixing father about 120V to be sure of that. And I misspoke, I meant "obsolete" more than "dead." I have access to a couple of CRTs and a 13" LCD that work, they're just completely useless.
|
|
#
?
Nov 21, 2018 01:35
|
|
|
No, my statement stands. With your scope, it's not the voltage that you have to worry about (although you do). It's the grounding situation that will arc through your scope, even on the low voltage side.
|
|
#
?
Nov 21, 2018 01:49
|
|
|
I'm sure there's a good reason why scopes don't float but gee whiz it is a trap for young players as EEV would put it.
|
|
#
?
Nov 21, 2018 04:56
|
|
|
I'm still shaking my head at my own stupidity. I have an amp I use on my PC and for the last few years it ha exhibited some weird issues with channels breaking up every now and again. I finally decided to take it apart and see if I can track it down. I had the cover off while playing some music and started tapping components and noticed that flexing the main board made the problem come and go. Dry solder joints perhaps, I'm thinking, but something else comes up in my life and I leave it for a few days. I come back to it and unscrew the board to flip it over, spying nothing too obvious but start to re-solder a few suspicious joints. I go to do some on a power transistor and BOOM, it flashes off and the power in my house drops out. How the gently caress did I forget to unplug the amp!? I go to turn the board back over and discover that one of the fuses has tack-welded itself to the amp chassis. Anyway, the end result is that the amp still powers on with a red LED lighting up, but that's it. A relay normally clicks over after a second or two for the speaker protection but now there's nothing. I have a bad feeling that 240v has been shown to a lot of important bits that don't like that sort of thing and I've totalled this thing beyond useful repair. drat gently caress poo poo, what an idiot thing to do, but at least I didn't die.
|
|
#
?
Nov 21, 2018 05:07
|
|
|
|
| # ? May 28, 2024 09:44 |
|
|
Eh, it happens to everyone eventually. I have a pair of flush-cutters with a pair of perfect little 1mm semicircular notches burnt into the blades -- I (attempted to) cut through the power wires going to a 3D printer's heater cartridge without turning off the power. 400 watts in a dead short through the blades.Shame Boy posted:Yeah, this old EEVBlog goes into why if you're interested: Dave is great but god loving dammit I wish he'd just pitch-shift his videos down about 15%
|
|
#
?
Nov 21, 2018 05:19
|
|