|
Break the ground pin off a power strip or use one of those adapters for old houses
|
#
?
Dec 17, 2019 02:15
|
|
|
|
| # ? May 9, 2024 05:51 |
|
|
CopperHound posted:Are you sure there isn't some center tap that isn't also tied to a common equipment ground? There is a center tap, for the other leg of the transformer. That one powers the output stage. It doesn't touch the ground anywhere.
|
|
#
?
Dec 17, 2019 03:00
|
|
|
I am trying to get any kind of .NET Core running on a Pi Zero, and running into issues I'd like to ping off somebody. Since it involves tiny computers, is this the right thread?
|
|
#
?
Dec 17, 2019 04:47
|
|
|
You might have better luck in the SH/SC Raspberry Pi thread, SH/SC Linux thread, or the Cavern of Cobol .Net thread depending on the exact nature of your problem.
|
|
#
?
Dec 17, 2019 04:58
|
|
|
babyeatingpsychopath posted:6.3VAC rectifies to 5.6VDC. This is the perfect spot for a linear regulator. Full-wave rectifier + smoothing cap (3000uF for 100mA draw) -> this thing's LDO linear regulator = win. How sure are you about that voltage? I got 7.24 VAC in for my rectifier and I'm getting 7.96 VDC out with a 3300uF cap and a 120 ohm load. Forgive me if I'm asking basic questions. I'm still learning how to build my own (tiny) power supplies and there's too many numbers involved for me to not stress out about them. I'm trying to understand all these settings on my oscilloscope too. I measured the above numbers with my DMM and I'm trying to figure out which unit terms coincide with it since its screen just says "volts". Does Vrms on the scope equal my DMM's units? It does say "True RMS Multimeter" across the top (Fluke 87V). Also, in the case of looking at DC power supply outputs, would ripple voltage be Vpp on the oscilloscope? How do you size the smoothing cap by the way? I experimented with different size spare caps I have. It was fun zooming in and seeing the ripples actually match diagrams online. It's so encouraging when your own screwing around actually matches examples! I figured out all by myself why a rectifier by itself outputs sine waves. They're like Goldilocks: the load can't be too big or too small. It has to be just right. I love looking at stuff with my oscilloscope by the way. It's shocking, no pun intended, to see how much stuff with electronics isn't precise, just "good enough". kid sinister fucked around with this message at 20:38 on Dec 17, 2019 |
|
#
?
Dec 17, 2019 20:19
|
|
|
kid sinister posted:How sure are you about that voltage? I got 7.24 VAC in for my rectifier and I'm getting 7.96 VDC out with a 3300uF cap and a 120 ohm load. Forgive me if I'm asking basic questions. I'm still learning how to build my own (tiny) power supplies and there's too many numbers involved for me to not stress out about them. Nah they're good questions, low-voltage AC being rectified is weird because you have to factor in a bunch of things like the voltage drop across the diodes in the bridge rectifier, the size of the cap, frequency of the AC, the load, etc. Unless you absolutely need to be more precise you generally aim for a range (like 6 to 10 volts) and just pick a regulator that can handle inputs between that range (with some margin for error at either end). As long as your regulator can handle that and your capacitor is rated for that voltage, I wouldn't worry about it. kid sinister posted:I'm trying to understand all these settings on my oscilloscope too. I measured the above numbers with my DMM and I'm trying to figure out which unit terms coincide with it since its screen just says "volts". Does Vrms on the scope equal my DMM's units? It does say "True RMS Multimeter" across the top (Fluke 87V). Also, in the case of looking at DC power supply outputs, would ripple voltage be Vpp on the oscilloscope? RMS (root mean square) is how AC voltage is measured, yeah. When your multimeter is set to AC voltage, it's actually measuring the AC RMS voltage, so Vrms on the oscilloscope should agree with that value. That only applies to AC though, DC is just measured as plain old volts on both the meter and the scope. Vpp is volts peak-to-peak, which if measuring a DC value would indeed be ripple. kid sinister posted:How do you size the smoothing cap by the way? I experimented with different size spare caps I have. It was fun zooming in and seeing the ripples actually match diagrams online. It's so encouraging when your own screwing around actually matches examples! There's calculations you can do to estimate the acceptable amount of ripple for your regulator and application and then pick the cap based on that, but I generally just throw a big cap in and hope for the best  The only problem you'll run into by doing that is you could blow a fuse if the capacitor is too big and tries to charge up instantly to the full voltage right when it turns on, but since it's on the other side of a transformer I'm not sure that's really much of a concern e: kid sinister posted:I figured out all by myself why a rectifier by itself outputs sine waves. They're like Goldilocks: the load can't be too big or too small. It has to be just right. A perfect bridge rectifier with nothing hooked up would create this wave:  However nothing's perfect, so real bridge rectifiers have leakage, forward voltage, nonlinearities, capacitive coupling between the AC and DC sides, and all sorts of other fun things that mess that ideal world up. When you add a load and/or a capacitor, most of that low-level stuff gets drowned out by the much higher current that's now flowing and the capacitor's smoothing, so it stops really being an issue. kid sinister posted:I love looking at stuff with my oscilloscope by the way. It's shocking, no pun intended, to see how much stuff with electronics isn't precise, just "good enough". Oh boy is it ever, the more you learn about electronics the more amazing it is that anything at all ever works reliably 
Shame Boy fucked around with this message at 21:06 on Dec 17, 2019 |
|
#
?
Dec 17, 2019 20:58
|
|
|
Even at work we almost never bother calculating the ripple cap size we need, you just take a punt based on experience. The only exception is for complicated DCDC power supplies, but they usually come with a formula in the spreadsheet that tells you what they need based on your load and permitted ripple current/voltage. But of course, both of those end up being "ehh good enough" values as well.
|
|
#
?
Dec 17, 2019 21:07
|
|
|
OK I'm doing something wrong. I got my power supply board up and working (thanks everyone!), but I'm getting a buzz when I attach audio output via a stripped headphone cable to the amp section of this radio. So I hook the shield of my audio cable to the radio frame and the dial lights dim even though they're LEDs and the I can't connect via Bluetooth to the Bluetooth board anymore. It starts getting hot too. It sounds like I made a circuit that shouldn't be... again. So I disconnected the audio output from the Bluetooth board and it survived. Still, how do I remove my hum? Which ground is the safe one to connect the shield to?
|
|
#
?
Dec 19, 2019 04:05
|
|
|
This is tough for me to give an answer without a solid grasp of all the stuff you are hooking up. I'm guessing we are dealing with a situation where your common voltage reference is not the same as earth ground. Try putting a volt meter between the ground of your Bluetooth thing and the radio chassis with that audio cable unplugged.
|
|
#
?
Dec 19, 2019 05:30
|
|
|
Basically, I took my AC power source from the heater line of the radio's own tubes, sent it through a rectifier and smoothing cap to power a Bluetooth board. I'm taking the audio from that board and feeding it back into the radio. I get 4 volts AC from the Bluetooth output ground to the chassis ground.
|
|
#
?
Dec 19, 2019 18:53
|
|
|
kid sinister posted:I get 4 volts AC from the Bluetooth output ground to the chassis ground.
|
|
#
?
Dec 19, 2019 19:01
|
|
|
CopperHound posted:You don't have access to a wiring diagram of the radio, do you? https://www.americanradiohistory.com/hd2/IDX-Service-&-Sales-IDX/Archive-Rider-IDX/IDX/40s/40/Rider-1940-PQR-OCR-Page-0052.pdf The 2 black leads from the main transformer power the heater lines and serve as chassis ground. I figured out the pinouts for the tubes if you'd like. They're not written on the schematic. kid sinister fucked around with this message at 19:12 on Dec 19, 2019 |
|
#
?
Dec 19, 2019 19:09
|
|
|
Okay. You can't share a ground between your ac source and full wave rectifier output. You end up creating a short for half the cycle. You can however share the ground with a half wave rectifier. Try drawing out the full wave as 4 diodes to imagine how the short happens.
|
|
#
?
Dec 19, 2019 20:33
|
|
|
I can imagine it. A half wave short was how I fried my first rectifier via my oscilloscope's ground. I had 2 probes, one hooked up in front and the second behind it at the same time.. So is this what I get for trying to be more efficient? I don't need a rectifier, just a diode of appropriate size?
|
|
#
?
Dec 19, 2019 21:15
|
|
|
Do you actually need to share a ground between the AC and your circuit? What's your chassis made from?
|
|
#
?
Dec 19, 2019 21:29
|
|
|
The chassis is steel. I pretty sure I do need to use the ground between them as that low voltage transformer tap is connected to the frame and used as the ground for everything that isn't the output stage. In other news, I'll need a regulator too as the preamp board I bought, despite being the same size from the same manufacturer, doesn't have one and needs 5 volts.
|
|
#
?
Dec 19, 2019 22:04
|
|
|
kid sinister posted:The chassis is steel. I pretty sure I do need to use the ground between them as that low voltage transformer tap is connected to the frame and used as the ground for everything that isn't the output stage. Yeah you're right, bugger that makes things tougher.
|
|
#
?
Dec 19, 2019 22:24
|
|
|
Splode posted:Yeah you're right, bugger that makes things tougher. Somehow I think this will involve grounding things at the radio end but not connecting them at the Bluetooth board end.
|
|
#
?
Dec 19, 2019 23:29
|
|
|
I guess that's why my professor said mixed signal people get paid the big bucks.
|
|
#
?
Dec 19, 2019 23:42
|
|
|
Any ideas for common electronics projects that would really benefit from artsy-fartsy metal cases/enclosures? I'm developing a novel Rapid Prototyping sheet metal forming workflow but have mostly done purely decorative sample parts so far; I'd like to exhibit a practical application where the technique's strong points are put to good use:
so far I'm considering: giving a generic rasp pi TV box a designer-rear end formed copper or brass enclosure that actually looks good in your living room; an SDR radio transceiver pi with bespoke RF shielding case and some kind of unusual form-factor for showing-off purposes; or a custom controller faceplate for a pi/arduino game emulator. they've all got fairly predictable dimensions, can be executed as clamshell-type two-half designs well-suited to the process, and are small enough dimension-wise to be tackled with my existing hydraulic press + other equipment. any other well-suited applications that would benefit from a bangin' fancy enclosure? Or do you maybe have a project or product under development that this sounds like the perfect answer for that I could run some prototypes for? 
Ambrose Burnside fucked around with this message at 00:23 on Dec 20, 2019 |
|
#
?
Dec 20, 2019 00:17
|
|
|
Cojawfee posted:I guess that's why my professor said mixed signal people get paid the big bucks. IDGI. Then again, I didn't go to college for this. drat. I can't test out my grounding theory since I can't power my preamp board to get nice, easy to use screw terminals since my power supply outs out too much voltage. The manufacturer's own page even says "Working voltage: DC 5V (please don't use more than 5V, or the IC will be burnt)". I'm getting like 7-8V out. I need a regulator. Which kind though? I'd imagine no switching since this is an AM radio. Wait a minute. If I'm just testing, I can throw together a voltage divider... Ambrose Burnside posted:any other well-suited applications that would benefit from a bangin' fancy enclosure? Or do you maybe have a project or product under development that this sounds like the perfect answer for? Tube amp! They're usually big old hunks of sheet metal for cases. How about a stove fan? They sit on top of wood stove to blow the rising hot air farther into the room. They're powered directly from the heat and for a power source use either a Sterling engine or a thermopile. I bet you could really creative with the heat sinks for the electric type! kid sinister fucked around with this message at 00:39 on Dec 20, 2019 |
|
#
?
Dec 20, 2019 00:21
|
|
|
kid sinister posted:Tube amp! They're usually big old hunks of sheet metal for cases. A tube amp fits into this as "larger goods where structural fabrication is done through conventional processes but is then Up-Marketed with custom-embossed hardware, fittings, badges and faceplates replacing mass-produced off-the-shelf hardware". Which, to be fair, is a legitimately-useful application of the process for manufacturers making high-end, small-run products. I've just already got a lot of Art Examples to show off, is all. I guess what I'm specifically looking for is 'single-process designs' where I can turn out near-finished small custom parts that would otherwise take several processes to manufacture, and are therefore overlooked by hobbyists or small shops with no specialized metal-forming equipment. It comes off as a strictly-decorative technique so I wanna emphasize that it has real and practical applications beyond art. For example, display panels/faceplates require shearing to form, punching and flanging of holes for buttons and display elements or ventilation louvers, engraving/printing/art transfer for graphics or lettering, etc- I can handle all of those with my process if the part is designed for this specific process, often in a single manufacturing step.
|
|
#
?
Dec 20, 2019 01:08
|
|
|
e: (apparently not an edit whoops) kid sinister posted:How about a stove fan? They sit on top of wood stove to blow the rising hot air farther into the room. They're powered directly from the heat and for a power source use either a Sterling engine or a thermopile. I bet you could really creative with the heat sinks for the electric type! I've actually got this near-exact project in my Ideas Book already as a purely-mechanical functional art piece/ showoff project, modeled after some antique examples of Stirling-powered fans from the 19th century I ran into. Uses a large, low-temp-diff gamma Stirling engine which will run without needing the drastic temp differentials of a woodburning stovetop. The fan head itself looks more like a wind turbine than a fan, with long+thin, low-pitch and lightweight blades that are actually repurposed RC glider wings; conventionally-shaped fans that move a lot of air per rotation are fine for stronger power sources, but you gotta min-max efficiency at every step with a low-differential engine. The thermal differential that makes it go is ideally near-passive, provided by evaporative water cooling on the cold cylinder and either solar or candle/lamp flame heat on the black-painted hot cylinder. The antiques I saw all used lamps in the pedestal base to provide a big differential, which is necessary for inefficient engines with a lot of friction in the piston/cylinder fit and the powertrain, but LTD Stirlings hadn't been invented yet, and now you can actually extract some modest useful work from, for example, the couple dozen degrees' diff between a sun-hot black metal cylinder and a water-soaked sponge in the wind. If you're up to machining your own self-lubricating piston heads from graphite to suit the exact dimensions of the piston cylinder, and spend hours and hours fine-tuning the assembly, anyways. People have run model engines off of a, like, 3 deg. C differential, I bet I could get some actual work out of ~25-50C's difference. My embossing process would only really shine for the old-timey brass trim and badging and for easy parts like the cold cylinder evaporation tray, the powertrain components generally need more precision than I can get from embossed parts. But yes, this is the sort of application I'm looking for, thanks much~ e: that's a shitload of words in this thread for something that does not require electricity to function, whoops Ambrose Burnside fucked around with this message at 01:58 on Dec 20, 2019 |
|
#
?
Dec 20, 2019 01:27
|
|
|
Cojawfee posted:I guess that's why my professor said mixed signal people get paid the big bucks. Nah, FPGA engineers and RF engineers get the big bucks. kid sinister posted:IDGI. Then again, I didn't go to college for this. I'm not sure what you're getting at with a voltage divider, as drawing literally any power from it will change the voltage division, so it won't work. I can tell you from experience Bluetooth devices' power draw varies wildly during operation. 7-8 to 5V is still a good use case for a linear regulator. Having a stash of TO220 package fixed 5 volt regulators is always handy for this sort of thing.
|
|
#
?
Dec 20, 2019 02:34
|
|
|
Splode posted:I'm not sure what you're getting at with a voltage divider, as drawing literally any power from it will change the voltage division, so it won't work. I can tell you from experience Bluetooth devices' power draw varies wildly during operation. Hey! It worked for all of 5 seconds. It was fun, watching the voltage on my DMM go down and its LED dimming... Yeah I need a regulator. I understand LDOs are for regulating when the supply voltage is close to the output. Is 2-3 volts off "close"? Or is "close" like a percentage of the input?
|
|
#
?
Dec 20, 2019 03:08
|
|
|
You can calculate that  https://www.analog.com/en/analog-dialogue/articles/understand-ldo-concepts.html
|
|
#
?
Dec 20, 2019 03:11
|
|
|
kid sinister posted:Hey! It worked for all of 5 seconds. It was fun, watching the voltage on my DMM go down and its LED dimming... “LDO” is poor term. Most “LDO” regulators are fine with inputs that are significantly higher than the output, often much higher.
|
|
#
?
Dec 20, 2019 03:44
|
|
|
LDO is Low Dropout, a measure of how close to the output the input voltage can be before the regulator stops functioning. Before LDO designs, they all needed something like 3V headroom. Before switchers became common, lots and lots of equipment had linear regulators turning like 12 volts or more into heat.
|
|
#
?
Dec 20, 2019 03:50
|
|
|
But they do get really loving hot if you go too high. Feel free to try, but be on the lookout for that symptom.
|
|
#
?
Dec 20, 2019 04:13
|
|
|
KnifeWrench posted:But they do get really loving hot if you go too high. Feel free to try, but be on the lookout for that symptom. I'm working with tubes. What's your definition of "really loving hot"?
|
|
#
?
Dec 20, 2019 04:17
|
|
|
Oh hell yeah they will, and it feels easy to make what seems like a reasonable increase in V or I and jump over the edge where the current cooling can't keep up and it gets hot really quickly. Modern ones have current and temperature limiting but silicon can handle getting way hotter than skin. For the OP: A few volts is no big deal unless the current is high. Even an amp is 1A * 3V = 3W, which would be really disappointing as a heater. If you look in the datasheet, you can find out how hot it will get at a certain power level and how hot it can get. taqueso fucked around with this message at 04:29 on Dec 20, 2019 |
|
#
?
Dec 20, 2019 04:26
|
|
|
Ambrose Burnside posted:Not a bad idea, but probably too large. Definitely should have mentioned that right now I can only run individual parts with a ~6"x6"ish envelope across the two biggest dimensions, and not from steel sheet or thicker gauges of nonferrous metals. Copper, brass, aluminium are more suited to the process, Throw some nixies in it and double your price.
|
|
#
?
Dec 20, 2019 04:46
|
|
|
kid sinister posted:I'm working with tubes. What's your definition of "really loving hot"? How much current do you need?
|
|
#
?
Dec 20, 2019 04:48
|
|
|
taqueso posted:How much current do you need? Hardly any. I'm basically powering 2 USB devices.
|
|
#
?
Dec 20, 2019 05:17
|
|
|
CopperHound posted:Steampunk cosplayers will line up to throw money at you for some custom stuff. I spent way too much time trying to make a working iris for goggles from some brass sheet for a Halloween party a few years ago. i hate that this is extremely shrewd + good thinking e: legit entirely unrelated- Is it possible to significantly restrict the FoV of one of those cheap thermopile arrays (like panasonic's grid-eye) through physical housing geometry without making the sensor useless for imaging purposes? 5-7 meters seems like the ballpark claimed range for effectively differentiating a human from their environment based solely on temperature using these sensors; for the application I'm considering I'd want something similar (albeit with medium-large animals instead of humans) to be possible out to about ~25 meters, and it's fine if the FoV is incredibly narrow. Imaging is the main application, but ideally it'd pull double-duty as a PIR-based motion detector that would wake the thing up and start the imaging, That 5-7m distance is quoted for sensors with bigass FoVs of ~60 degrees; there's a specialized narrow-focus grid-eye with a 35-degree FoV, but I don't think that shrinks the imaged area enough at the distances I'm needing. So- it seems too simple to work, but can I just stick one of these inside a tube or cone-shaped housing that occludes everything outside of, i dunno, a 5-10-degree FoV, maybe? Or could you use conventional reflecting lenses to alter the FoV in a controlled way, maybe by cannibalizing a cheap telescope kit or something? (I know conventional glass focusing lenses block IR but just mirrored lenses should transmit the needed wavelengths, i think) If I actually have to build a reflecting telescope to make this work it's obviously time to consider a different type of sensor, but I want an excuse to screw around with one of these thermopile arrays, so I'm willing to make my life a little harder than it has to be here. Ambrose Burnside fucked around with this message at 06:51 on Dec 20, 2019 |
|
#
?
Dec 20, 2019 06:45
|
|
|
Ambrose Burnside posted:Any ideas for common electronics projects that would really benefit from artsy-fartsy metal cases/enclosures? I'm developing a novel Rapid Prototyping sheet metal forming workflow but have mostly done purely decorative sample parts so far; I'd like to exhibit a practical application where the technique's strong points are put to good use: Try the smaller schools who won't have access to metallized 3d printing.
|
|
#
?
Dec 20, 2019 13:50
|
|
|
VictualSquid posted:Might be nice for some crazy experimental antennas, depending on the tolerances. Back in grad school, I knew several people who worked on strange organically shaped antennas and who would have loved a better manufacturing process. That's a great idea, if I can find someone doing that sort of thing- applications like that are even better than enclosures, the more purely functional/utilitarian the better- the decorative/aesthetic potential is obvious and already well-demonstrated, I want to highlight the more concrete design/prototyping/manufacturing challenges it can help with. Another part of this workflow I'm playing with actually does something very similar. I'm working with two rapid tooling techniques with different niches- one that covers large, low-detail designs made with extremely cheap and quickly-made tools, and another for very high-detail, high-precision designs where the tooling isn't so convenient or cheap but is still potentially-accessible to the well-equipped hobbyist or prototyping shop with no CNC mill at hand. The overall idea is to combine the two techniques prudently to produce composite tools for a wide range of designs as quickly and cheaply as possible. The fine detail process is electroforming, basically electroplating a very thick plating of metal (copper in my case) over a conductive paint layer applied to a model, which is later removed from the plating to produce a "shell" of the tool that's then filled with incompressible cast epoxy resin to produce a sturdy final product good for up to hundreds or thousands of part cycles. Electroforming is a black art of sorts, though, apparently very difficult to master; one of the most common issues is inconsistent plating thickness, which leads to weak tools, pitting/surface defects,etc. And the only way you can tweak that thickness in a targeted way during the plating process is through the cathode geometry + location- the closer it is to the part, the more metal is deposited. Anyways- to make high-quality cathodes that plate evenly, I'm gonna feed the first (low-precision, quick n dirty) process into the second and use it to emboss copper sheet cathodes that evenly and consistently conform to the model/tool being electroformed, so all parts will get plated with an approximate even distance from the depositing element. Not that huge an application, but it feels satisfyingly- elegant. Ambrose Burnside fucked around with this message at 18:42 on Dec 20, 2019 |
|
#
?
Dec 20, 2019 18:31
|
|
|
God drat. If it isn't one thing, it's another. I already burnt up 2 rectifiers for God knows what reason, probably a short. Then when I finally got my power supply board built and putting out a nice, clean 5V, its output freaks the gently caress out when I plug in this preamp board. The preamp is fried. A replacement is already on the way. Why can't anything just work?
|
|
#
?
Dec 21, 2019 02:03
|
|
|
you should get a circuit breaker or fuse in there somewhere post the wiring you have going on exactly as it is now / how it was when the pre amp died
|
|
#
?
Dec 21, 2019 02:06
|
|
|
|
| # ? May 9, 2024 05:51 |
|
|
taqueso posted:you should get a circuit breaker or fuse in there somewhere gently caress, I didn't even think of a fuse, and now my board is too compact to put one where there should be. All I got on hand is an in line holder with 10 gauge wire for 1.25"x.25" fuses. I do have some 250V 1A fuses for it too. I got a feeling this is total overkill though. The 10 gauge wire won't even fit in my screw terminals... I can't prove the preamp ever worked to be honest. This Bluetooth board I can attach to it works just fine though. kid sinister fucked around with this message at 02:20 on Dec 21, 2019 |
|
#
?
Dec 21, 2019 02:12
|
|