|
Discussion time! How do you think they made the head on the Mini MagPrinter? http://www.popularmechanics.com/technology/gadgets/reviews/a10696/this-is-awesome-printing-your-own-magnets-16900288/ https://www.youtube.com/watch?v=drD416THU7Y Obviously an electromagnet of some kind, but shape? Type of wire? How many millions of amps are they putting through that thing?
|
#
?
Jul 24, 2015 08:27
|
|
|
|
| # ? Jun 8, 2024 00:55 |
|
|
I have a quick question. I'm trying to do some DIY shielding on an existing power cable and a signal cable. Is there more to it than just wrapping it in aluminum foil and wiring the foil to a ground? Can I wrap the foil around the plastic jacket of an existing cable? I have a switching power supply that seems to be interfering with a signal and I'd like to avoid spending a lot of money on new cables.
|
|
#
?
Jul 24, 2015 17:36
|
|
|
Alfred P. Pseudonym posted:I have a quick question. I'm trying to do some DIY shielding on an existing power cable and a signal cable. Is there more to it than just wrapping it in aluminum foil and wiring the foil to a ground? Can I wrap the foil around the plastic jacket of an existing cable? I have a switching power supply that seems to be interfering with a signal and I'd like to avoid spending a lot of money on new cables. That's a really good question, and I'm curious to find out the answer. It's possible that mylar from a baloon would be easier and cheaper to work with if flexibility of the cable is necessary.
|
|
#
?
Jul 24, 2015 17:47
|
|
|
ante posted:Discussion time! I was about to make some educated guesses about how that system works, but I deleted everything I wrote and just looked up the patents instead. However, now I'm temporarily out of time, so feel free to look into it yourself: https://www.google.com/patents/US20140211360 and for more patents/applications from that company (since the patent office search engine is flakey AF): http://patents.justia.com/assignee/correlated-magnetics-research-llc
|
|
#
?
Jul 24, 2015 18:21
|
|
|
never read a patent
|
|
#
?
Jul 24, 2015 19:45
|
|
|
Alfred P. Pseudonym posted:I have a quick question. I'm trying to do some DIY shielding on an existing power cable and a signal cable. Is there more to it than just wrapping it in aluminum foil and wiring the foil to a ground? Can I wrap the foil around the plastic jacket of an existing cable? I have a switching power supply that seems to be interfering with a signal and I'd like to avoid spending a lot of money on new cables. The real solution is to get the braid designed for making shielded cable assemblies, but i'm guessing that it's pretty expensive (only used it at work), and then you're looking at redoing the cabling. Copper tape is probably easier to work with, or at a minimum some aluminium tape instead of normal foil. The copper tape can be soldered, which makes it the better choice, but might be harder to find locally. Depending on the circuit (and assuming it helps) you might need to terminate one or both ends to chassis, ideally both ends should be terminated in a properly engineered system, but with DIY addons it can make things worse.
|
|
#
?
Jul 24, 2015 22:19
|
|
|
JawnV6 posted:never read a patent But the abstract starts with this hilarious sentence: Magnet Man posted:A magnetizer for magnetizing magnetic field sources into a magnetizable material includes a magnetization subsystem, a motion control system, and a magnetizer control system. MAGNETS MAGNETIZING MAGNETIC MAGNETS MAGNETICALLY
|
|
#
?
Jul 24, 2015 22:23
|
|
|
But how do they work
|
|
#
?
Jul 24, 2015 23:24
|
|
|
Patent clerks lying to us and getting us pissed.
|
|
#
?
Jul 24, 2015 23:26
|
|
|
If I had a device that transmitted serial data, like say a GPS board, and I wanted it to spit data out to two devices simultaneously -- I understand that I can't just wire the Tx of the sender to both Rx of my receivers. Is there any other way to do this? This communication would be strictly one way so I don't necessarily need or want communication from either receiver to the original transmitter.
|
|
#
?
Jul 26, 2015 04:56
|
|
|
Martytoof posted:If I had a device that transmitted serial data, like say a GPS board, and I wanted it to spit data out to two devices simultaneously -- I understand that I can't just wire the Tx of the sender to both Rx of my receivers. Is there any other way to do this? This communication would be strictly one way so I don't necessarily need or want communication from either receiver to the original transmitter. Actually, unless there are issues keeping the TX and both RX devices on a common ground, this should probably work. Although the TX can only drive so much current, it is probably more than enough for several RX devices. I don't think the UART RX devices will have pull ups, so that shouldn't be an issue.
|
|
#
?
Jul 26, 2015 05:14
|
|
|
Martytoof posted:If I had a device that transmitted serial data, like say a GPS board, and I wanted it to spit data out to two devices simultaneously -- I understand that I can't just wire the Tx of the sender to both Rx of my receivers. Is there any other way to do this? This communication would be strictly one way so I don't necessarily need or want communication from either receiver to the original transmitter. You can totally do this assuming the data rate is not that high. Run the Tx through a buffer like the 74xx125 or 126 and that will help with potential integrity issues.
|
|
#
?
Jul 26, 2015 08:54
|
|
|
Thanks all for the soldering iron tip tips. I went and purchased some (1 USD/piece) Friday night, but they are ever so slightly larger than what I have.  Back to the electronic store I guess.I have a few questions about some components. First - Interface cables I'm using 6 simple M/F jumper wires for my 6 segment display. I'd like to locate it about 4 feet from the device. One option is to build the Pi into the enclosure, but the digits are about 2 centimeters shorter than the pi in any orientation. The device has a 12 VDC power input. If I build a 12V to 5v stepdown converter using an LM317, I can get this down to 5 signal cables. I believe the total mA draw of my device is quite small since I'm using ULN2008s. I do wonder if the last digits will lose sync. What are people using for more "permanent" devices between their Pi/Arduino and pcbs. I know the solution for a lot of Arduino projects is just to put a Atmel 328 right on the PCB, but I'm still prototyping/programming in Python first before transitioning to an Arduino that is maybe controlled sent data wirelessly via zigbee. The best solution I can come up with is to either 1) continue using m/f jumper wires 2) use a bulky DB9 cable 3) use Cat5. Second - Enclosures I would like to sandwich this display between two pieces of transparent material. I read I need a diffuser for 7 segment displays. My options seem to be: glass (oh no), polycarbonate/lexan, or acrylic. I need (2) 8 x 60 cm pieces. This covers the front of the digits and backs of the PCBs. It certainly won't be IPxxx standard compliant, but what should I do fro the top and bottom, (2) 3 x 60, and "end", sections? I suppose I should also use standoffs to mount the pcbs into sandwich material to make it stronger. Third - LM324N Quad Amplifiers I took apart my 67.2 Volt electric bike battery carger to fix it and found this chip at the heart of it. What is a good resource for learning about battery chargers? I found this through googling, but the circuit in my charger seems a little bit more complex. http://www.circuitsgallery.com/2014/01/battery-charge-controller-circuit.html Eventually I would like make a wireless charger. Is there a good primer out there for wireless power transmission?
|
|
#
?
Jul 26, 2015 11:46
|
|
|
Slanderer posted:Actually, unless there are issues keeping the TX and both RX devices on a common ground, this should probably work. Although the TX can only drive so much current, it is probably more than enough for several RX devices. I don't think the UART RX devices will have pull ups, so that shouldn't be an issue. Inputs should be very high impedance on almost everything made in the last few decades (since CMOS became en vogue) so the drive current shouldn't be an issue. Line length and the resulting capacitance can potentially be an issue for rise and fall times but it likely won't be a problem if it's all on board and at normal UART speeds. If it's all stuff on different boards try to keep the wires short.
|
|
#
?
Jul 26, 2015 22:11
|
|
|
movax posted:You can totally do this assuming the data rate is not that high. Run the Tx through a buffer like the 74xx125 or 126 and that will help with potential integrity issues. Or split right at the driver with individual source series termination.
|
|
#
?
Jul 26, 2015 22:27
|
|
|
politicorific posted:Third - LM324N Quad Amplifiers If this is an Li-Ion battery, I would strongly recommend not diving into it, ESPECIALLY for a 60+ volt charger. This is both in the realm of power electronics and li-ion batteries, which means that mistakes can lead to fires and serious injury. The chip you found is just a standard quad op-amp. Op-amps are the heart of an enormous range of analog circuit topologies, so anything could be going on in there. The chip is probably being used for multiple, separate functions at the same time (it is cheaper and saves space to use multiple op-amps in the same package in a lot of situations, with individual sub-circuits using different op-amps on the chip) BattleMaster posted:Inputs should be very high impedance on almost everything made in the last few decades (since CMOS became en vogue) so the drive current shouldn't be an issue. Yeah, I was considering mentioning the capacitance, but I think it's less of a concern than with I2C (where you have the pullup to form an RC filter with). FAKE EDIT: Looking it up, the maximum cable length for RS232 @ 19200 baud (for example) is 50 feet, compared to only a couple meters (typically) for I2C Slanderer fucked around with this message at 01:36 on Jul 27, 2015 |
|
#
?
Jul 27, 2015 01:27
|
|
|
Hmm, ok that's neat. I'm basically looking to get telemetry out of an RC receiver/transmitter module. It outputs the telemetry to the control radio but I'm hoping to siphon that off to a Raspberry Pi or generic SD datalogger so I can import my flights into Splunk later. Sounds like it may be doable.
|
|
#
?
Jul 27, 2015 04:24
|
|
|
Slanderer posted:Yeah, I was considering mentioning the capacitance, but I think it's less of a concern than with I2C (where you have the pullup to form an RC filter with). Yeah, pull-ups are terribly slow (though the only way to have a multi-drop output without worrying about having to put things in high-Z mode) while the UART is likely to be directly-driven by a CMOS output which is significantly faster. As you said it probably isn't an issue so it should be no big deal to just split it off.
|
|
#
?
Jul 27, 2015 04:29
|
|
|
On RC receivers/transmitters, are there any small modules that are good for a simple detect signal -> pin high deal? I have a pair of nRF24L01 modules but looking at the datasheet it doesn't seem as though I could do something like that without a microcontroller in between.
|
|
#
?
Jul 27, 2015 04:31
|
|
|
Parts Kit posted:On RC receivers/transmitters, are there any small modules that are good for a simple detect signal -> pin high deal? I have a pair of nRF24L01 modules but looking at the datasheet it doesn't seem as though I could do something like that without a microcontroller in between. Not RC stuff, but instead standard remote control tech. What you what are simple transmitter/receiver pairs with built in encoder and decoder. While you can use simple RF modules with an encoder+decoder, doing so means that anything else on your transmitting frequency will be received (like garage door openers, for instance). Both the transmitter and receiver have an address/code set by DIP switches (or a permanent method). Sparkfun sells a few modules (did they have more at some point?) but without encoding. Seeedstudio also sells some(definitely not FCC approved lol), including a kit with encoder/decoder. Here is another retailer reselling it since Seeedstudio may have discontinued it: http://garagelabstore.com/store/communication/2km-long-range-rf-link-kits-w-encoder-and-decoder.html This can technically be done with RC transmitters and receivers too, but those will be more expensive (since they have much more complicated radio ICs). I honestly can't remember if a standard RC transmitter will send a DC signal, or if it will only send PWM signals. In the latter case, you would need a pulse encoder on one end (likely a 555 circuit) and a pulse decoder on the other (probably just an RC filtered comparator) Slanderer fucked around with this message at 18:11 on Jul 27, 2015 |
|
#
?
Jul 27, 2015 18:06
|
|
|
Slanderer posted:Not RC stuff, but instead standard remote control tech. What you what are simple transmitter/receiver pairs with built in encoder and decoder. While you can use simple RF modules with an encoder+decoder, doing so means that anything else on your transmitting frequency will be received (like garage door openers, for instance). Both the transmitter and receiver have an address/code set by DIP switches (or a permanent method). Going to second this. I've used those modules (or an incredibly similar one) and they're really easy to use. They're all basically RF carrier boards for the PT 2262 / PT 2272 encoder decoder pair. You'll usually find it on 433 or 315 mhz, though there are circuits for use on IR as well. That one says 2km range. Using the cheapest ones we could find, we never got more than a couple of hundred meters out of them. I don't know how much of it was quality, and how much of it was our application. (behind metal shielding + low to the ground. hilariously bad)
|
|
#
?
Jul 27, 2015 19:11
|
|
|
This guy has done a nice comparison of those Nrf modules http://youtu.be/gtM832Z0ujE
|
|
#
?
Jul 27, 2015 23:21
|
|
|
Will less lovely potentiometers give me more stable, consistent readings when read by an Arduino's analogue pin? I'm having a hell of a time debouncing the cheap one I bought (tried the capacitor between analog and ground, and a couple of software solutions).
|
|
#
?
Jul 29, 2015 02:59
|
|
|
Audio pots will be very clean. But if time isn't critical, why don't you just use an even larger capacitor? That's pretty much what they're for.
|
|
#
?
Jul 29, 2015 05:00
|
|
|
rawrr posted:Will less lovely potentiometers give me more stable, consistent readings when read by an Arduino's analogue pin? I'm having a hell of a time debouncing the cheap one I bought (tried the capacitor between analog and ground, and a couple of software solutions). Wouldn't you want to average (low pass filter) the signal, rather than debounce? I'm not sure what your application is, are you trying to use the pot like a switch? Are the readings not stable when the knob is not being turned? What are you doing to process the data on the arduino now?
|
|
#
?
Jul 29, 2015 05:29
|
|
|
The Arduino has what, a 10-bit ADC? You won't really see noise beyond the baseline you'd get with shorted inputs until it's in the millivolt range, so if you're seeing a significant amount you're probably coupling in noise from elsewhere.
|
|
#
?
Jul 29, 2015 05:37
|
|
|
taqueso posted:Wouldn't you want to average (low pass filter) the signal, rather than debounce? I'm not sure what your application is, are you trying to use the pot like a switch? Are the readings not stable when the knob is not being turned? What are you doing to process the data on the arduino now? Thanks for mentioning the low pass filter - that's the term I needed to search for instead of debounce. http://playground.arduino.cc/Main/Smooth seems to be working great! e: My issue was with the values moving around even if the knob is left untouched, and values jumping around (i.e. 1,2,3,4,5,4,5,6).
|
|
#
?
Jul 29, 2015 05:55
|
|
|
rawrr posted:Thanks for mentioning the low pass filter - that's the term I needed to search for instead of debounce. http://playground.arduino.cc/Main/Smooth seems to be working great! If you're getting that much of a stability issue, something's probably awry with your potentiometer. Once you've left the potentiometer untouched for a few seconds, you shouldn't see much more bouncing around than back and forth between two adjacent values unless something's mechanically or electrically unstable in your circuit.
|
|
#
?
Jul 29, 2015 06:14
|
|
|
Looking at making up a quick power regulator for a tiny camera module to run it off the same power supply as a 12v car monitor. So since vehicle voltage can go up to 14ish volts, let's say 14.5 to be safe, and the rated draw for the camera is 70 milliamps if I'm understanding the heat dissipation calculation for the little regulator correctly I should expect (14.5-5)v x 0.070amps = 0.665 watts. Should I be okay to use this without a heat sink attached to the voltage regulator?
|
|
#
?
Jul 31, 2015 11:37
|
|
|
Parts Kit posted:Looking at making up a quick power regulator for a tiny camera module to run it off the same power supply as a 12v car monitor. So since vehicle voltage can go up to 14ish volts, let's say 14.5 to be safe, and the rated draw for the camera is 70 milliamps if I'm understanding the heat dissipation calculation for the little regulator correctly I should expect (14.5-5)v x 0.070amps = 0.665 watts. Should I be okay to use this without a heat sink attached to the voltage regulator? If it's something like a 7805 regulator in a standard package, I think it should be about to do 0.66W without a heatsink, but I don't actually have the specs on hand. It might get hot though, so it'll probably need to be enclosed. The bigger issue is actually the car's 12V supply. If there's any chance you might have this plugged into the car while it's running (or when it's started), then the regulator will see spikes of a much higher voltage (50-100V or something) which may destroy it. The DC input will need to be clamped and/or LC-filtered, I think. I've never had to do this before personally, but it's been discussed in this thread before so someone should have more specific advice. Alternatively, you can probably buy a 5V buck supply with car DC filtering/protection for $5. Maybe less fun than building the supply yourself, but quicker to get this working. One last thing---if you're hooking this into your car's DC, there needs to be an inline fuse (basically all cigarette lighter plugs have a fuse inside them, and car monitors have a fuse inline on the power cable.
|
|
#
?
Jul 31, 2015 12:45
|
|
|
Slanderer posted:If it's something like a 7805 regulator in a standard package, I think it should be about to do 0.66W without a heatsink, but I don't actually have the specs on hand. It might get hot though, so it'll probably need to be enclosed. This may not be helpful but I don't buy the 50-100v spikes. I need to see a scope trace on a modern car to buy it. For one the battery is a strong stable source that can source or absorb 100's of amps. I also don't believe that cheap convinient store accessories are rated for these things and they do ok. And I see zero reason why a modern car designer would tolerate such spikes. It's cheaper to prevent them than to design 100 loads to individually absorb them. But maybe I don't know what I'm talking. As for temperature double check the junction to ambient thermal rise. 0.66W is ok for TO-220 which are typically like 50ish degrees per watt.
|
|
#
?
Jul 31, 2015 17:06
|
|
|
At the moment it'll just be on a bench, so everything should be at 12V max once I unbox my power supply, which would give just under a half a watt dissipation. I'm not 100% sure I'll be using this camera connected to the car since it's really for another project (crappy nightvision goggles) and a 12v monitor is just what I have on hand to test it with, but if I do it I will keep the possibility of spikes in mind until I figure something else. It might just end up being used to help me monitor what's going on in the engine compartment while I do other poo poo anyways so that's good to know. On the fuse -- yeah the monitor has an inline fuse already so I'm good on that front. On spikes -- would that be the sort of thing a zener diode setup would deal with? Like say find one with a rating of 14-15v and put that & a beefy resistor to ground so any overvoltage doesn't go to the regulator in the first place?
|
|
#
?
Jul 31, 2015 17:38
|
|
|
asdf32 posted:This may not be helpful but I don't buy the 50-100v spikes. I need to see a scope trace on a modern car to buy it. For one the battery is a strong stable source that can source or absorb 100's of amps. Here is a document about automotive electronics design; Chapter 1 provides an overview of some common issues.
|
|
#
?
Jul 31, 2015 18:41
|
|
|
asdf32 posted:This may not be helpful but I don't buy the 50-100v spikes. I need to see a scope trace on a modern car to buy it. For one the battery is a strong stable source that can source or absorb 100's of amps. A couple links: https://en.wikipedia.org/wiki/Load_dump http://www.littelfuse.com/data/en/application_notes/an9312.pdf That app note is from 1999, but I don't think much has changed since then. It all comes down to the alternator, and any poo poo that generates is going right into the 12V electrical system. Bad brushes on the alternator, a bad battery connection, a jump start---all these can cause spikes. On top of that, you have the shitload of noise from fans, AC compressor, and anything else electrical. I spoke with our lead EE (since we have a few devices with cigarette jack power cables and over engineered safety circuits), and he confirms that this is all an issue. However, he says that with a 7805 and big capacitors, you'll probably survive common transients just fine. The noise will still be an issue, but that might not be a huge problem for Part Kit's application
|
|
#
?
Jul 31, 2015 19:42
|
|
|
Just spend a few cents more on an automotive-rated linear regulator.
|
|
#
?
Jul 31, 2015 19:58
|
|
|
asdf32 posted:This may not be helpful but I don't buy the 50-100v spikes. I need to see a scope trace on a modern car to buy it. For one the battery is a strong stable source that can source or absorb 100's of amps. For what it's worth, the LM2937 which is a automotive grade regulator, (~1.60 on digikey in to-220) indicates that it's protected "up to +60 V/ −50 V load dump transients." It and the LM2940 regulators have been recommended here before. One of the automotive switcher modules I've seen was rated to only 40V transients. Either way, it will tell you a bit about what they expect these to need to survive.
|
|
#
?
Jul 31, 2015 20:03
|
|
|
asdf32 posted:This may not be helpful but I don't buy the 50-100v spikes. I need to see a scope trace on a modern car to buy it. For one the battery is a strong stable source that can source or absorb 100's of amps. Trust me, they're rated for them. Car utility sockets are basically Satan's voltage source. Spikes of +/- 30V are basically your normal running situation; alternators generate huge momentary voltage transients. Even with everything working perfectly, you'll see spikes of -50V or so when you turn a vehicle off, as the engine reaches 0rpm and the alternator's inductive field collapses entirely. (You also need to gracefully deal with brownout voltages -- i.e. you can expect even solid batteries to drop to 6V or below while cranking the engine.) Generally, the worst condition you can run into is alternator load dump -- that's where, due to corrosion or other mechanical failure, the battery gets entirely disconnected while the vehicle's in motion, and the alternator is the sole source of current for the vehicle. Without the battery in the circuit, you will absolutely see 100V spikes. I can't find it at the moment, but I remember reading an application note for some regulator that was designed for use in LED taillights, and its block diagram was incredibly complicated; some international standard said that taillights had to absorb spikes between +100V and -300V without permanent damage.
|
|
#
?
Aug 1, 2015 02:27
|
|
|
Slanderer posted:A couple links: Ah well a battery disconnect while running is one thing I wasn't considering (heh I should have, I'm in the middle of designing a high power inductive discharge module for a different application). The Littelfuse appnote is excellent (I like their stuff) but it doesn't quite tell us what to expect at a cigarette outlet because it seems there will be some suppression in the distribution: quote:Protection by a Central Suppressor
|
|
#
?
Aug 1, 2015 03:51
|
|
|
I guess all this stuff doesn't matter right now anyways since the little TFT I got off amazon was DOA and has to be returned. 
|
|
#
?
Aug 1, 2015 04:43
|
|
|
|
| # ? Jun 8, 2024 00:55 |
|
|
Now I'm wondering how hybrids and modern cars with auto stop/start electrical systems are like. The electrical system manages to start and stop the motor with no noticeable effect on the car electronics many times per journey, and for hybrids at least I'd expect more low level high frequency noise from the DC/DC converters, but all the aux electrics run off the 12V battery while the engine start and ignition presumably run off the high voltage pack. Never popped the hood on the new VWs with auto stop start, but I'm guessing they maybe have a smaller aux battery for the car electronics to prevent brownouts every time you let go of the brake?
|
|
#
?
Aug 1, 2015 12:38
|
|