|
taqueso posted:I believe you can 're-foam' the woofers but it's just something I've read about in passing. I did a pair of Paradigm bookshelf speaker drivers a few months ago and it went fine. I used a kit from Simply Speakers purchased through Amazon. Simply Speakers has great instructions and videos and I wouldn't hesitate to do it again. 
|
#
?
Mar 19, 2017 18:17
|
|
|
|
| # ? May 20, 2024 08:58 |
|
|
taqueso posted:I believe you can 're-foam' the woofers but it's just something I've read about in passing. Yeah I'm aware of that and actually have a kit waiting to install on my main set. I just don't feel like it's worth it for 4 mystery woofers of unknown and probably poor quality.
|
|
#
?
Mar 20, 2017 00:16
|
|
|
asdf32 posted:And what are you using for control out of curiosity? Digital or analog control IC's? I'll be doing all the control in the FPGA which is work but means I know I can deal with dead time etc (though with finite timing resolution). quote:Yeah I'm having a hard time seeing how EPC fits into my current requirements but I'm still staring at the Transphorm stuff which are basically drop in replacements for regular devices up to 600V. We have to do a couple PFC designs and they have some interesting "Totem Pole" PFC demo boards.
|
|
#
?
Mar 20, 2017 00:39
|
|
|
I'm rebuilding two crossovers for a pair of old JBL PA speakers. I need to replace a power resistor marked (Ph)10w16rj and a 4 microfarad 100V 5% cap. I assume the resistor is 10 watt, 16 Ohm, but I'm not sure that's what the 16rj is indicating. Also the cap appears to be a film and foil cap, but they are quite expensive compared to a metalized polypropylene cap. ($36 vs $1 on partsexpress. Can anyone chime in on the resistor marking and also whether or not the cheaper cap will likely sound much different?
|
|
#
?
Mar 20, 2017 13:49
|
|
|
There are so many marking systems, it's hard to be sure, but I think you're right. The 'J' appears to be a 5% tolerance: http://www.engineeringtoolbox.com/resistors-value-letter-digit-code-d_1656.html Capacitors have many other values besides capacitance. Effective series resistance (ESR) tends to be the one that can matter that isn't obvious, especially when filtering. I couldn't tell you how to expect the qualitative experience to change, and the cost difference may be enough to try regardless, but don't be shocked if it isn't a drop-in replacement.
|
|
#
?
Mar 20, 2017 14:32
|
|
|
rawrr posted:https://www.aliexpress.com/wholesale?catId=0&initiative_id=SB_20170314172221&SearchText=fc+idc Oh, duh, ribbon cable connectors. I didn't think of that. Thanks.
|
|
#
?
Mar 20, 2017 21:46
|
|
|
Question on powering DC motors: I have an application at work where I need to power a series wound DC motor. Not typical in my line of work. A question has risen about diodes. I have a flyback diode spec'd in my design, but the power supply manufacturer has a vague recommendation for a blocking diode: http://www.magna-power.com/support/technical-notes/getting-best-performance-ac-dc-power-supplies-abusive-environments Scroll down to the bottom where they have sections on "back fed voltage" and "reverse voltage." I've been told by others in my company that they have used a blocking diode based on this "spec sheet," but I don't see where it's necessary (though that is likely due to my ignorance). In my line of work, we use flyback diodes all the time for relays and other inductive loads in order to prolong the life of the switching relays. Using a blocking diode is new to me, and we only have 1 other time we've had a similar application so I don't see that as confirmation either way. So the questions: 1) Is a blocking diode more appropriate for this, or is a flyback diode correct? 2) Would a blocking diode be in addition to, or in place of, a flyback diode? 3) Dow do I know which type of diode I need for a given application? This is like a $15k power supply so I'd be very interested in not destroying it.
|
|
#
?
Mar 21, 2017 00:20
|
|
|
ANIME AKBAR posted:For something like a multilevel inverter I would definitely be using an FPGA, but most of my work is on more simple bridge converters for amplifiers. For that I need to operate with a really wide duty cycle range so I make discrete PWM modulators. One of the huge issues I've had is that the LM5113 has a silicon bug where putting in a pulse with a width of 3-4ns will cause both outputs to go high for much longer, which usually makes things go boom. So my modulators are carefully designed to make sure that never happens. Most control ICs can't make that guarantee though. So custom high speed analog modulators? That's neat. Well that brings me to another question, how many watts you can you effectively dissipate in those little EPC devices? Seeing 16C/W to board for one part so just a couple watts realistically I assume.
|
|
#
?
Mar 21, 2017 03:55
|
|
|
DaveSauce posted:Question on powering DC motors: Pretty sure you want both, unless you need a very fast switching off of the motor/are super concerned about cost, in which case you should think more. You don't really need the blocking diode, but it gives you some piece of mind. The way I read it, the back fed voltage part is concerned about (1) having a battery or cap or something that will hold the V+ rail up when the supply is off and (2) having something that will hold it above it's nominal output. Either way something is pushing current back into it and potentially damaging things. The blocking diode stops that from happening. The reverse voltage part is saying that most of their supplies have an internal diode that will conduct if you apply reverse voltage, but it might not be big enough for whatever abuse you're applying (or it might not exist). The peak voltage the blocking diode would be exposed while switching off the motor with no flyback would be large and hard to figure out. The blocking diode should have high enough amperage ratings to handle your peak load + margin, a high enough voltage rating to block whatever is likely to get backfed (at least nominal output voltage + some), and low enough resistance to not waste a bunch of power.
|
|
#
?
Mar 21, 2017 05:41
|
|
|
Foxfire_ posted:Pretty sure you want both, unless you need a very fast switching off of the motor/are super concerned about cost, in which case you should think more. You don't really need the blocking diode, but it gives you some piece of mind. Thanks, I think I have a much better picture of it now. I think my main concern was if there would be some reason NOT to include a flyback diode aside from redundancy. My application can probably handle an additional diode cost-wise, but our biggest constraint is space. Loaded running current for the motor is on the order of 200A, so I've got some nice big wire going around. That said, it's a $200 diode, and if that saves a $15k power supply I'm all for it. Here's a follow-up question: What would the current consumption of the blocking diode be? Part of what we're doing is testing the current consumption of the motor, with our measurement target being +/- 0.5 A. We have a shunt that does this, but if the diode throws off our measurement we may be in trouble. Basic question I know, but it's been a long time since I worked at the component level like this. As a reference, this is the diode we are using for the flyback: http://www.pwrx.com/pwrx/docs/qrs061k001.pdf Is there any reason not to use that for a blocking diode as well?
|
|
#
?
Mar 21, 2017 13:22
|
|
|
asdf32 posted:Well that brings me to another question, how many watts you can you effectively dissipate in those little EPC devices? Seeing 16C/W to board for one part so just a couple watts realistically I assume.
|
|
#
?
Mar 22, 2017 13:38
|
|
|
I have an HDPlex DC-DC power supply which claims an input range of 16-24v, 19v nominal: http://www.hd-plex.com/HDPLEX-300W-Hi-Fi-DC-ATX-Power-Supply-16V-24V-Wide-Range-Voltage-Input.html I am trying to pair it with a Meanwell power supply, HRP-600: http://www.meanwellusa.com/webapp/product/search.aspx?prod=HRP-600 There are multiple versions of that model, one with a 15v nominal output and one with a 24v nominal output. They can both be adjusted by potentiometer, and the ranges are: 15v (adjusts to 13.5 ~ 18V) 24v (adjusts to 21.6 ~ 28.8V) My question is, if I can't hit the 19v nominal input of the HDPlex, is it better to put in a couple more volts or a couple less? I just want the best efficiencies out of this combination, and the lowest chance of anything blowing up because it's running my PC.
|
|
#
?
Mar 22, 2017 22:13
|
|
|
Zero VGS posted:I have an HDPlex DC-DC power supply which claims an input range of 16-24v, 19v nominal: Sounds like you're micro-optimising your decision. On paper the 15v nominal, adjusted to 16v would be ideal. Only exception is if the wire length from the ACDC PSU to ATX DC-DC converter is over, say, 10ft and thin in diameter - in which case going in at the top end (24v output) is most efficient to avoid losses through resistance in the wiring. Either way it's much of a muchness unless you're constantly drawing 300w. And if that's the case, I'd want to consider some active cooling on the HDPlex, depending where you put it. Jamsta fucked around with this message at 22:37 on Mar 22, 2017 |
|
#
?
Mar 22, 2017 22:35
|
|
|
I want a central controller to communicate over some wires to smaller microcontrollers elsewhere in a project (3-4 of them, preferably all using the same bus to not use too many pins on the master controller). I was thinking i'd use I2C since I've used it for other stuff but I forgot it was designed specifically to be a local bus rather than to operate over wires. The wires shouldn't be more than 3-4ft long and will have an external foil shield but in addition to the data lines the wires will have power, which will probably spike a bunch due to switching stepper motors on and off. Is this a bad idea? Is there some other bus meant to do what I'm trying to do? I know RS-422 and RS-485 are designed to be more noise-immune but they seem like pretty big overkill for what I'm doing... My other idea was to just slap an nRF24L01 on each controller since I bought a bunch of cheap ones off ebay a while back that I've been itching to use in something, but that also seems kinda, uh, ridiculous 
|
|
#
?
Mar 23, 2017 19:41
|
|
|
It shouldn't be hard to get that to work. If you have extra wires use twisted pairs and wrap each i2c line around ground. Or seperate i2c from the power. Use stronger pull-ups to reduce bus impedance, like 1k. There are also options to bridge i2c to other physical layers. See PCA9615.
|
|
#
?
Mar 23, 2017 19:58
|
|
|
I²C is good for 6 m on paper. I think you’re fine. Apply capacitors and inductors liberally.
|
|
#
?
Mar 23, 2017 20:01
|
|
|
ate all the Oreos posted:RS-485 ate all the Oreos posted:nRF24L01
|
|
#
?
Mar 23, 2017 20:05
|
|
|
JawnV6 posted:do you need more than a level shifter/inverter for this? It's just "uart, but -5 to +5 and inverted" right?
|
|
#
?
Mar 23, 2017 20:07
|
|
|
JawnV6 posted:how hard are your realtime requirements? IIRC, you're given a time slice that can be arbitrarily interrupted by the BLE stack, if you can't handle more than 30ms latency/jitter it's not a great option Wait is the nRF24L01 actually based on BLE? I thought it was its own weird thing. (I assume BLE is bluetooth low energy and not something i'm missing right?) e: I forgot to actually answer the question - I don't think 30ms latency would be that big a deal, it's basically just sending "do [thing] and report back when you're done" asynchronous messages so it doesn't need constant feedback or very tight coupling between the pieces asdf32 posted:It shouldn't be hard to get that to work. If you have extra wires use twisted pairs and wrap each i2c line around ground. Or seperate i2c from the power. Use stronger pull-ups to reduce bus impedance, like 1k. Platystemon posted:I²C is good for 6 m on paper. Ah okay, thanks. I probably should have just tried it experimentally first to see if it would actually be a problem before asking but I've been designing the thing and writing code for it for the last like 10 days and it just now hit me that "oh hey maybe you should see if it works first" and I got worried 
Shame Boy fucked around with this message at 20:31 on Mar 23, 2017 |
|
#
?
Mar 23, 2017 20:25
|
|
|
ate all the Oreos posted:Wait is the nRF24L01 actually based on BLE? I thought it was its own weird thing. (I assume BLE is bluetooth low energy and not something i'm missing right?) Ah, sorry, didn't check the datasheet. I'd assumed it was one of their SDR SoC's with a blob you drop in to specify which flavor of 2.4G transmission including BLE.
|
|
#
?
Mar 23, 2017 20:28
|
|
|
I2C's range is limited by line capacitance. The stated maximum is 400 pF which I think comes from a maximum allowable rise time at the maximum clock speed and using the recommended pull-up resistor value. You can probably get away with more capacitance by using a lower clock speed and/or a lower-value pull-up (being careful not to exceed the current sinking rating of the line drivers) but 400 pF should be pretty hard to hit unless you do something crazy.
|
|
#
?
Mar 23, 2017 20:32
|
|
|
I'm planning out a new instrument cluster/on-board computer for my motorcycle, and I am puzzling over some theoretical questions about data storage that maybe some people in here have an opinion on. So, the system is going to have a speedometer, tachometer, temperature gauge, etc. and all of that is straightforward enough. It also needs to have a primary odometer to keep track of overall miles, a couple of trip odometers, and likely some other types of persistent data about the bike. I aim to store this stuff in EEPROM. The odometers need to have resolution to at least a tenth of a mile, and I don't want to lose any of it when the bike shuts down...but 10 writes per mile means 100,000 writes in 10,000 miles, which means most kinds of nonvolatile storage will burn out in less than two years of riding for me. So I have a couple of different options here. 1) Switch to a replaceable storage medium, like an SD card. Not a great idea because it's slow to write and doesn't really solve the problem. Might still use this for other forms of data-logging down the line, though. 2) Build in some sort of brownout-detection system that only writes when the bike is shutting down. With a decently sized capacitor on the microcontroller power lines, I could keep the chip running for a few seconds after the power is cut, and the chip could monitor its own voltage and only store the data when it goes below a threshold. An advantage here is that I don't need to tie up the chip for a few milliseconds every tenth of a mile -- I'm aiming to do everything from screen updating to engine RPM pulse-counting with the single device, if possible. 3) Attempt some kind of wear-leveling. The microcontroller I'm aiming to use, a Teensy 3.2, has a 72MHz Cortex-M4 with 4k of EEPROM. That's plenty of space to move the data around in the storage, theoretically; storing let's say 16 bytes of persistent data and pushing it forwards every 1000 miles, I would be able to get ~250,000 miles before looping around, and I could do that ten times before approaching the ~100k write wear limit. I'm leaning towards the last option, but it kind of depends on how quickly the data can be stored. I haven't timed it exactly but specs suggest on the order of 1ms. At the 9500RPM redline, I have about 80 tachometer pulses per second to deal with, which is about 12ms of space between interrupts -- so I don't think this should be an issue but it remains to be seen. I also want to update the screen at 30Hz, read from a speedometer wheel sensor at about the same rate, and do a bunch of sensor handling and data processing. If I can't get acceptable performance, option #2 might be better. What's the best choice here? Are there any other options? How are digital odometers handled in modern cars? Sagebrush fucked around with this message at 22:01 on Mar 23, 2017 |
|
#
?
Mar 23, 2017 21:59
|
|
|
Microchip sells a new type of EEPROM that they call EERAM. EERAM stores writes in SRAM during normal operation. When it detects a power loss it writes the SRAM contents to EEPROM. When it's powered back up it restores the SRAM from the EEPROM. The end result is that it only writes to the EEPROM once per power cycle leading to much less wear for constantly-overwritten data. This is essentially your #2 but it would be less work for you to implement yourself. http://www.microchip.com/design-centers/memory/serial-eeram
|
|
#
?
Mar 23, 2017 22:04
|
|
|
ate all the Oreos posted:
Sup https://hackaday.com/2017/02/08/taking-the-leap-off-board-an-introduction-to-i2c-over-long-wires/ BattleMaster posted:Microchip sells a new type of EEPROM that they call EERAM. EERAM stores writes in SRAM during normal operation. When it detects a power loss it writes the SRAM contents to EEPROM. When it's powered back up it restores the SRAM from the EEPROM. The end result is that it only writes to the EEPROM once per power cycle leading to much less wear for constantly-overwritten data. Woah
|
|
#
?
Mar 24, 2017 01:34
|
|
|
nvSRAM is the generic name for those kind of chips if you're hunting for parts on digikey
|
|
#
?
Mar 24, 2017 04:14
|
|
|
I thought NVSRAM was when you've got SRAM backed up by a battery, not this particular configuration. edit: I mean NVSRAM would be just fine for the application but this EERAM stuff is pretty cheap and is more compact BattleMaster fucked around with this message at 04:44 on Mar 24, 2017 |
|
#
?
Mar 24, 2017 04:39
|
|
|
That's actually one of the use cases for the new generation of exotic storage, like FRAM. Not saying you should go that route (you'd need to read datasheets as I bet there's not a prefab Arduino library for them), but this is the sort of market segment the weird non-volatiles are carving out.
|
|
#
?
Mar 24, 2017 04:45
|
|
|
Stabby McDamage posted:That's actually one of the use cases for the new generation of exotic storage, like FRAM. Not saying you should go that route (you'd need to read datasheets as I bet there's not a prefab Arduino library for them), but this is the sort of market segment the weird non-volatiles are carving out. I bought some FRAM to play around with just the other day (hasn't arrived yet) and as far as i can tell they're designed to be drop-in replacements for parallel SRAM, and i assume the ones that have I2C / SPI interfaces work similarly 
|
|
#
?
Mar 24, 2017 04:48
|
|
|
BattleMaster posted:I thought NVSRAM was when you've got SRAM backed up by a battery, not this particular configuration. It covers all the types. External cap to power it, external battery to power it, and internal battery in the IC
|
|
#
?
Mar 24, 2017 05:09
|
|
|
Anyone know where I can find a cheaper alternative to this: https://www.sparkfun.com/products/13740 I can only find it on the sparkfun page, didn't see it on aliexpress or similar. Low power, small needed for a diy data logger
|
|
#
?
Mar 24, 2017 14:40
|
|
|
A Yolo Wizard posted:Anyone know where I can find a cheaper alternative to this: https://www.sparkfun.com/products/13740 This may be what you want, note it's the Ublox 6 not the 7 - so it's functional spec is different. Still does GPS over serial nicely. http://www.ebay.com/itm/Ublox-NEO-6...kMAAOSwhOVXdieB
|
|
#
?
Mar 24, 2017 14:58
|
|
|
ate all the Oreos posted:I bought some FRAM to play around with just the other day (hasn't arrived yet) and as far as i can tell they're designed to be drop-in replacements for parallel SRAM, and i assume the ones that have I2C / SPI interfaces work similarly Cypress I2C FRAMs have the same interface as standard EEPROMs. There's potentially some differences in timing since writes complete much faster than flash based systems, but that would just give slower write performance unless the driver was modified. For industrial applications another difference worth mentioning is the data retention at high temperatures is dramatically lower than flash (151 years at 65 degrees C, but only 10 years at 85 degrees).
|
|
#
?
Mar 25, 2017 11:42
|
|
|
I like wiring electric guitars and ever since I discovered the whole logarithmic sounds pressure relation I decided to use only A potentiometers. But there's this doubt always creeping in my mind: does it change things wether I solder the output to the right side or the left side? (Input being soldered on the swipe?). Or, if I am soldering the output onto the central lug, will the log taper be wrong if I lug it to one or the other side? Thanks!
|
|
#
?
Mar 25, 2017 12:17
|
|
|
Dawncloack posted:I like wiring electric guitars and ever since I discovered the whole logarithmic sounds pressure relation I decided to use only A potentiometers. But there's this doubt always creeping in my mind: does it change things wether I solder the output to the right side or the left side? (Input being soldered on the swipe?). This determines whether clockwise or counter clockwise is "more". That's probably reason enough to get it right for a guitar. Log pots are actually piecewise linear (http://www.resistorguide.com/potentiometer-taper/), so wiring it one way or the other will put the "steep" section at low or high resistance, which does matter. Which one you want depends on the rest of your circuit and the desired behavior.
|
|
#
?
Mar 25, 2017 14:36
|
|
|
Ok, so they don't work symmetrically. Now that I know time to run some tests! Thanks, for some reason google didn't help and it was eating me.
|
|
#
?
Mar 25, 2017 16:59
|
|
|
I bought a cheap Chinese digital clock kit off eBay because I figured it'd be a good way to get better at soldering and have a neat project to show for it. Solder all the components, plug it in, everything works, even the 7-segment display with the pin that needed to be fixed! Great! Put together the little laser cut acrylic case for it, put it in the case, no longer turns on 
|
|
#
?
Mar 25, 2017 17:25
|
|
|
Sockser posted:I bought a cheap Chinese digital clock kit off eBay because I figured it'd be a good way to get better at soldering and have a neat project to show for it. Solder all the components, plug it in, everything works, even the 7-segment display with the pin that needed to be fixed! Great! Reflow all your joints, check for shorts. Poke it a bunch to check for intermittent connections
|
|
#
?
Mar 25, 2017 18:03
|
|
|
ante posted:Reflow all your joints, check for shorts. Poke it a bunch to check for intermittent connections Easier said than done. All the components are on one side of the board, and then the four display panels are on the other side, covering up the joints from the first side. Pain in the rear end. It's nearly identical to this: https://www.youtube.com/watch?v=8Ty5Bnj2XqE e: update: had to grab something out of the basement, decided to take a quick look at it and see if anything was obviously wrong. Nothing visible to my eyes. Plugged it in, now it works. Problem solved, I guess. Sockser fucked around with this message at 18:37 on Mar 25, 2017 |
|
#
?
Mar 25, 2017 18:17
|
|
|
Hi there friends. I've basically never touched anything electronic, but I'm trying to spin a motor using a rpi 3 and an L298N motor controller. As a first step, I've just tried to get the L298N to turn on. Following directions I found on the internet, I have taken a 12V power supply and cut the end off. I stripped the wires, found the wire with the dashed lines on it which should be + according to reading I've done and hooked that up to the 12v+ thing (what is the name for these? terminals?). I've then hooked the other wire, along with a jumper cable to a ground on my pi, up the the ground... thing. Then I plugged the power supply into the wall, and nothing happened. Near as I can tell, this controller should have a red LED lighting up when it is powered up. What are my troubleshooting steps here?  
|
|
#
?
Mar 26, 2017 21:16
|
|
|
|
| # ? May 20, 2024 08:58 |
|
|
The driver board needs a 5v supply to power its own logic. The 12v line you've connected just powers the motors. I see on the datasheet that the board also has its own 12v-to-5v regulator that can be activated by removing the jumper (a small plastic block stuck on two pins) above the 12v power lines. Have you tried that? If that doesn't work, it looks like you'll have to provide a regulated 5v supply onto the third pin, marked in your datasheet as connected to 5v VCC from the microcontroller. Connect a wire from a 5v line on the raspberry pi (if it has one, I can't remember) to the third pad next to the 12v lines, as indicated. In order to spin the motors you'll also need to send a signal from the Pi's GPIOs, but one step at a time.
|
|
#
?
Mar 26, 2017 21:42
|
|