|
The ad was deleted now I bought it, but IIRC it has got an RS-232 port on the back
|
#
?
Sep 30, 2021 13:53
|
|
|
|
| # ? Jun 11, 2024 09:53 |
|
|
SNR theory question for everyone. I've asked this around in some other places and always seem to get different answers. In the figure below, my signal is "broadband" and occupies bandwidth BW. Fn is my Nyquist frequency. https://imgur.com/a/4nKFGma SNR is typically defined as signal power over noise power, but it seems to me there are 3 different geometric interpretations for defining the integration limits (which, for my toy example, just becomes areas-of-rectangles). Case #2 is treating everything within BW as signal, which is definitely not correct, so I think we can rule out this one. Case #1 seems reasonable, but you certainly could have a signal power less than the noise floor, which this interpretation doesn't allow. Case #3 seems reasonable, but its a bit weird in that it looks like you'd be double-counting that overlapped region in the signal power and noise power. So EE goons, which one would you choose?
|
|
#
?
Oct 5, 2021 17:17
|
|
|
iirc, case 3 is the correct one. Noise power is the power you have with the signal turned off; Signal power is the power you get with noise turned off.
|
|
#
?
Oct 5, 2021 17:31
|
|
|
How can I tell if a K-type thermocouple plug/jack is legit (i.e. made with the correct alloys)? I blanched at the cost on Digikey so bought the most reputable set I could find off Amazon, trusting in the reviews, but the different build quality and colour/appearance of the plug and jack components are making me suspicious. I’ve never used one of these before so I have nothing to go off of, but the plug pins look to be made of the same alloy, or at least have an identical appearance, which doesn’t sound right to me.
|
|
#
?
Oct 7, 2021 03:34
|
|
|
Have you got an x-ray spectrometer?
|
|
#
?
Oct 7, 2021 04:07
|
|
|
More practically, if the thermocouple behaves plausibly for your altitude in an ice water bath and in boiling water, it's probably good enough for hobby purposes. Sideways-ly, unless you need the wide operating temperature range, a thermistor is going to be cheaper, easier to integrate, and more accurate, especially if you're only building one device and can do a per-part calibration (the error is mostly in amount of material and properties of that batch, it'll be stable for the same physical thermistor). If you won't go hotter than 250C or so, a 5% part is less than a dollar.
|
|
#
?
Oct 7, 2021 04:25
|
|
|
How do we feel about NTCs
|
|
#
?
Oct 7, 2021 04:30
|
|
|
Ambrose Burnside posted:How can I tell if a K-type thermocouple plug/jack is legit (i.e. made with the correct alloys)? I blanched at the cost on Digikey so bought the most reputable set I could find off Amazon, trusting in the reviews, but the different build quality and colour/appearance of the plug and jack components are making me suspicious. I’ve never used one of these before so I have nothing to go off of, but the plug pins look to be made of the same alloy, or at least have an identical appearance, which doesn’t sound right to me. Put it in icewater. Put it in boiling water. If you get 0+-0.5 and 100+-0.5 C send it. Make sure you hook it up right. If you get it hotter and your sensor sez the temperature is going down, then it's backwards. I got a gotcha last time I got thermocouple jacks: I ordered K-type, and received J-type.
|
|
#
?
Oct 7, 2021 04:37
|
|
|
ante posted:How do we feel about NTCs NTCs are super cheap and good enough for most uses, they're pretty much the default in the HVAC industry from what I can tell for example. AFAIK the issue with NTCs is a combination of accuracy, and linearity vs. temperature that means they might not be great for something expecting a wide range of temperatures. The issue is that they have a large change in resistance vs. temperature over a relatively narrow temperature range, so you get great resolution around that temperature range, but outside that range the resolution of your measurement system goes down. In the old days they were also annoying to deal with in digital systems, but if you have a floating point unit in the processor (or enough ROM to store software emulation routines) it's really not a big deal to linearize them.
|
|
#
?
Oct 7, 2021 19:36
|
|
|
While we're having temp sensor chat, if you want to be really cheap, ordinary diodes forward biased with a current source (or a resistor) make good temperature sensors. The range is inconvenient if you don't have a high resolution ADC though, so you can't always get away with it. Here's a red LED (usually you'd use a small signal diode but you can use anything that behaves like a PN junction) with a 1MΩ load on 9V at room (so, about 7.5μA). then at soldering iron hot:  then at upside-down-air-duster cold:  You can buy already-calibrated chips that do this internally and provide you with a nice i2c or spi interface that spits out temperature digitally, but if your goal is temperature-sensor-at-lowest-cost and you already have a good ADC, maybe keep it in mind. The load is arbitrary but less power means less self-heating. This thing is barely glowing but you could definitely use an indicator LED as a temperature sensor.
|
|
#
?
Oct 7, 2021 20:44
|
|
|
You gotta be careful to keep a consistent light source with that, though. Otherwise you're measuring lux and not temp. A very confusing troubleshooting scenario
|
|
#
?
Oct 7, 2021 21:55
|
|
|
Yeah LEDs work both ways. Hitting them with light will generate current.
|
|
#
?
Oct 7, 2021 22:04
|
|
|
Good point. You'd want to use a current that was large compared to the photocurrent. I think most indicator LEDs are already there when they're lit (mA vs nA, I think?), but mine here with only a few μA is going to be more light-sensitive.
|
|
#
?
Oct 7, 2021 22:09
|
|
|
Cross-posting from 3D printer thread, I'm organizing my work area and have a lot of strips of SMD components. Anyone have a good way to organize these? Some of them are too long for binders. Does anyone know where to get the binder pages with the long slots for this though, regardless? Seen a good 3D-printed organizer?
|
|
#
?
Oct 7, 2021 23:28
|
|
|
longview posted:NTCs are super cheap and good enough for most uses, they're pretty much the default in the HVAC industry from what I can tell for example. RTDs are the only linear one, thermocouples aren't linear either. Thermocouples do have the simplest measuring circuitry; you can do a decent job with a single ended ADC, high accuracy thermistor/RTD want differential ADC and an external reference. poll plane variant posted:Cross-posting from 3D printer thread, I'm organizing my work area and have a lot of strips of SMD components. Anyone have a good way to organize these? Some of them are too long for binders. Does anyone know where to get the binder pages with the long slots for this though, regardless? Seen a good 3D-printed organizer?
|
|
#
?
Oct 8, 2021 04:20
|
|
|
fwiw i’m building a tempering/burnout oven, wherei’d regularly be operating above the usual thermistor range, otherwise i’d just go with that.
|
|
#
?
Oct 8, 2021 10:11
|
|
|
So is an ESR meter a waste of time and money? I figured it'd be a cool tool for checking electrolytic caps without desoldering. I am being told by one guy they on a more local non-english electronics forum that they are unreliable and bad practice and might even cause damage to the circuit in some cases. From what I know it's an older guy who has worked with electronics on a high level. He tells me to instead check visually first, then use a scope and check larger parts of circuits that they perform as they should, look for "ripples" or I guess noise and such that can indicate problems. Desolder and check components individually if testing is needed. OTOH I see loads of videos talking about how great they are even by Dave from EEVblog, I see loads of people on various forums talk about how often they use theirs and that they repair electronics for a living. I just hate it when I don't know a subject well and I meet people who have hard opinions and then their opinions seem to conflict with others. Who is right? I was thinking of building this one (because it's cool and retro) but now I don't know: http://kripton2035.free.fr/analog%20esr/esr-poptronix.html
|
|
#
?
Oct 8, 2021 12:52
|
|
|
You straight up can't check components while they're in circuit. That's because they'll be in parallel with the rest of the circuit, so your results will be meaningless
|
|
#
?
Oct 8, 2021 15:36
|
|
|
I don't know why, in a world with LCR meters to check component values and oscilloscopes to check ripple, you'd want a single-tasker like that unless you were a shop that specialized in servicing things with expensive caps that were known to fail by developing high ESR. It has the same feel as buying a tomato slicer for your kitchen: if you have to ask you probably don't need it yet.
|
|
#
?
Oct 8, 2021 16:04
|
|
|
The way people have been saying they use them, is to check if an electrolytic capacitor is good or bad without desoldering it. Which sounded really useful to me, since I am looking at some audio equipment where the caps might be going bad and I also read that they can sometimes go bad without obvious signs like bulging. If there are better ways to detect stuff like that, like using an oscilloscope to check the circuit for some telltale signs (that'd be ripple I guess) then I'd be all for learning that.
|
|
#
?
Oct 8, 2021 16:14
|
|
|
His Divine Shadow posted:The way people have been saying they use them, is to check if an electrolytic capacitor is good or bad without desoldering it. Which sounded really useful to me, since I am looking at some audio equipment where the caps might be going bad and I also read that they can sometimes go bad without obvious signs like bulging. As other people said, you can't really check for ESR in-circuit in general because the rest of the circuit is gonna throw it off. I say "in general" because some circuits behave better than others - if you only ever work on vacuum tube stuff, there's a good chance you can just measure stuff like ESR in circuit because when a vacuum tube is off it's basically an open circuit anyway and isolates that node from the rest of the device. But you have to have knowledge of the rest of the circuit to be able to tell that the cap is in fact isolated by the tubes, so it's not exactly the kind of technique you start with even if you're working on vacuum tube stuff. Looking for ripple when it's powered on is a pretty good idea, since that's the primary symptom of a bad cap that actually matters to the rest of the circuit. At the very least you could narrow down which caps you need to de-solder and check that way. Obligatory warning to be real real careful if it's wall powered and not isolated, though.
|
|
#
?
Oct 8, 2021 16:25
|
|
|
Also for audio repair stuff the guy I trust the most is Mr. Carlson. Like his actual job is repairing amplifiers for musicians, so he's been doing this for decades. He's a bit of an insane perfectionist that spends way too much time and effort making sure every single part of everything he repairs is perfect, but he gives reasons and theories behind everything he does which is super helpful. He's also got a dedicated full electronics repair course available if you donate to his Patreon so that's neat.
|
|
#
?
Oct 8, 2021 16:32
|
|
|
Shame Boy posted:Also for audio repair stuff the guy I trust the most is Mr. Carlson. Like his actual job is repairing amplifiers for musicians, so he's been doing this for decades. He's a bit of an insane perfectionist that spends way too much time and effort making sure every single part of everything he repairs is perfect, but he gives reasons and theories behind everything he does which is super helpful. He's also got a dedicated full electronics repair course available if you donate to his Patreon so that's neat. Huh, I didn't know that's what his job was, that's pretty awesome. But yeah, his videos are very, very good. You gave a spot on description of the content of his videos, but I'd just add that his perfectionism I think is an asset to his videos. They are very well produced with clear audio and well set up shots. They can be a bit of an information overload because he can get into such low level details and they're usually pretty drat long, but they're still great. I'll often throw on one of his hours long restoration videos just for nice background noise or to fall asleep to.
|
|
#
?
Oct 8, 2021 16:56
|
|
|
His Divine Shadow posted:The way people have been saying they use them, is to check if an electrolytic capacitor is good or bad without desoldering it. Which sounded really useful to me, since I am looking at some audio equipment where the caps might be going bad and I also read that they can sometimes go bad without obvious signs like bulging. I restore old radios. I don't even bother checking the electrolytic caps for 2 reasons: 1. At this age, they're most likely bad, and 2. Caps are cheap. You are right in that they can go bad without obvious signs. The worst is if they short internally. If you don't want the headache in the future, spend a little extra and replace them with some 10k-20k hour rated caps. I've also replaced some lower Farad-rated electrolytic caps with ceramics so I'll never have to worry about them again. They do make dedicated capacitor testers. The problem with them is that you need to take the capacitor out of circuit. Since Mr. Carlson has already been brought up, he does have plans oh his Patreon for a dedicated capacitor checker if you're interested. Hold old of equipment are we talking here? In the past, they liked to use funny values for Farads and axial caps were a lot more common than today.
|
|
#
?
Oct 8, 2021 21:01
|
|
|
Ambrose Burnside posted:fwiw i’m building a tempering/burnout oven, wherei’d regularly be operating above the usual thermistor range, otherwise i’d just go with that. If air temperature variation across your shop is small compared to your process control needs, you could assume the device pcba temp is close enough to the thermocouple cold junction temp and not worry about having thermocouple wire all the way back to the pcba
|
|
#
?
Oct 8, 2021 21:10
|
|
|
kid sinister posted:I restore old radios. I don't even bother checking the electrolytic caps for 2 reasons: 1. At this age, they're most likely bad, and 2. Caps are cheap. You are right in that they can go bad without obvious signs. The worst is if they short internally. If you don't want the headache in the future, spend a little extra and replace them with some 10k-20k hour rated caps. I've also replaced some lower Farad-rated electrolytic caps with ceramics so I'll never have to worry about them again. Mostly 70s and 80s for now, nothing tube driven edit: Unrelated but I found a brush pen for fountain pen ink in a drawer that was unused, I filled it with solder flux and hey I got a nice little solder brush pen. Looks like this, it'll be interesting to see if the flux adversely affects the pen over time 
His Divine Shadow fucked around with this message at 09:54 on Oct 9, 2021 |
|
#
?
Oct 9, 2021 06:02
|
|
|
His Divine Shadow posted:Mostly 70s and 80s for now, nothing tube driven It may. I had one of those that I put flux in and the alcohol dissolved the plastic and the whole thing turned milky white, then fell apart.
|
|
#
?
Oct 9, 2021 18:33
|
|
|
Looks like the MG Chemicals 835-P. It's my favourite flux.
|
|
#
?
Oct 9, 2021 18:49
|
|
|
Hey thread, sorry I've been out of touch but here's a question. Are there devices specifically designed to attach to surface mounted switches in order to override them, so to speak? I'm asking because I saw this: https://www.etsy.com/listing/979132858/pocket-operator-midi-adapter-v2?ref=shop_home_active_1&crt=1 ... which, to cut to the chase, you lay over the button panel of a pocket operator and it clamps on, connecting with all 16 buttons, and it sends open and close signals that a MIDI adapter gives it. I really thought that a MIDI adapter would be too much of a PITA but this guy has done it. I want to do something to a similarly minimal device, an NTS-1, supplementing all its little buttons with larger buttons that are easier to press and can really take repetitive use, in this case arcade buttons. It can already receive MIDI but it would be nice to be able to hit those switches fast. so do I need to manually solder them or is there a ready-made device for no-fuss clamping onto a standard-issue surface-mounted switch? Thanks again!
|
|
#
?
Oct 10, 2021 15:52
|
|
|
According to the description, it uses pogo pins (spring loaded pins) to come in contact with exposed traces on either side of the switch. I assume it has its own switch between the pogo pins, which then bypasses the original switch. In order to do that yourself, you would need exposed traces next to whatever switch you want to bypass.
|
|
#
?
Oct 10, 2021 16:08
|
|
|
Cojawfee posted:According to the description, it uses pogo pins (spring loaded pins) to come in contact with exposed traces on either side of the switch. I assume it has its own switch between the pogo pins, which then bypasses the original switch. In order to do that yourself, you would need exposed traces next to whatever switch you want to bypass. Pogo pins, eh? Doesn't sound as ready-made as I'd hoped. And here are the buttons in question RN. 
|
|
#
?
Oct 10, 2021 16:19
|
|
|
Is that exposed enough? And ideally I'd like to replace those little knobs with footpedals. But the buttons would be a great start. I really feel like this thing is the bomb and that making its functions a little easier to access will make a difference. I'm seeing various spring loaded contacts at Mouser, having a search term to work with now, but not sure I've found my el dorado yet.
|
|
#
?
Oct 10, 2021 16:45
|
|
|
Those buttons aren't surface mount, they are through hole. I looked at an assembly video for that kit, and you can see the button leads on the other side of the board. Aside from some sort of physical mechanism that pushes the actual button, you'd have to interface with the buttons from the underside.
|
|
#
?
Oct 10, 2021 16:51
|
|
|
Cojawfee posted:Those buttons aren't surface mount, they are through hole. I looked at an assembly video for that kit, and you can see the button leads on the other side of the board. Aside from some sort of physical mechanism that pushes the actual button, you'd have to interface with the buttons from the underside. Oh jeez I'm so sorry, I jumped to a conclusion there. Well let me see what the backside of this looks like then, this doesn't necessarily stop this. ED: In fact, it looks like it makes it easier, and you can access the knobs via the back also, so I might be able to, for instance, replace one of them with a pedal. I can't see too well what we have here but it looks pretty straightforward: 
petit choux fucked around with this message at 17:39 on Oct 10, 2021 |
|
#
?
Oct 10, 2021 17:28
|
|
|
petit choux posted:Oh jeez I'm so sorry, I jumped to a conclusion there. Well let me see what the backside of this looks like then, this doesn't necessarily stop this. If they're through-hole and you have access to it, and you don't modifying the thing, then unsolder the knob and replace it with a pedal. With the knob in place, there's a very good chance that your pedal won't do what you think it will, because it can't "replace" the knob. Resistors in parallel and all that. Your pedal can do the math to make the response something predictable and linear (or log or whatever) if knob to a fixed value, but if you're tweaking the knobs then that affects the pedal's effect on the input the knob is controlling. If that's fine, then just solder to the knob. The knob has three pins, +, -, and signal. Signal in these circuits is USUALLY voltage-based, so just having a pedal that can source or sink enough current to get that voltage to be whatever you want will work, within the limits of the original circuit's ability to source and sink current, i.e. if the knob is set to a "very low" voltage, and you try to drag it down further, it's possible that either your pedal can't "suck" that voltage away enough, or that the original circuit will try to "push" voltage into your pedal beyond its capabilities and burn itself out. This is a simplistic description, but I think it gets the gist. Others will be along with some math if you'd like. If these knobs are controlling current, then the job is way, way easier, as current sources and sinks play pretty well together and your pedal can sit in parallel with the knobs. If the knobs are affecting some other value (inductance, capacitance) then it gets really tricky, but it doesn't look like those are variable inductors or capacitors. little bit of followup: Since buttons are open-circuit in most implementations, connecting another open-circuit in parallel then closing it is precisely equivalent to "controlling the switch" since the base circuit can't tell the difference between your circuit closing and its button being pressed. "Analog" values are different, and require different techniques.
|
|
#
?
Oct 10, 2021 18:17
|
|
|
babyeatingpsychopath posted:If they're through-hole and you have access to it, and you don't modifying the thing, then unsolder the knob and replace it with a pedal. With the knob in place, there's a very good chance that your pedal won't do what you think it will, because it can't "replace" the knob. Resistors in parallel and all that. Your pedal can do the math to make the response something predictable and linear (or log or whatever) if knob to a fixed value, but if you're tweaking the knobs then that affects the pedal's effect on the input the knob is controlling. If that's fine, then just solder to the knob. Yeah I think running two pots in parallel might get pretty interesting but Imma keep it simple this go-round. I can start with buttons and of course keep the option of doing the pots later. But it's a really fragile little case and needs something more suitable if you're going to be rooting around under there, those tiny screws can chew away the threads they just cut real easily and then there you are aren't you. I happen to have tons of high quality standoffs so I'd replace all the corner pieces with plain standoffs and make them stand higher so one could access the underside more easily. I'd run wires out to a separate panel that has a bunch of arcade buttons, something like that. Eventually I'll design a box.
|
|
#
?
Oct 10, 2021 19:10
|
|
|
I’m back on my diy lavalier mic kick. I should be able to wire this to a USB-c connector, right? I am 99% sure I can, I just can’t find a good pin diagram for a USB-c.
|
|
#
?
Oct 11, 2021 06:53
|
|
|
Marsupial Ape posted:I’m back on my diy lavalier mic kick. I should be able to wire this to a USB-c connector, right? I am 99% sure I can, I just can’t find a good pin diagram for a USB-c. Use of USB-C for analog audio is called Audio Adapter Accessory Mode and I found this application note from TI that has some data on it including this which may be helpful: 
|
|
#
?
Oct 11, 2021 07:05
|
|
|
BattleMaster posted:Use of USB-C for analog audio is called Audio Adapter Accessory Mode and I found this application note from TI that has some data on it including this which may be helpful: I’m not great at reading schematics, yet. I guess it won’t be as simple as wiring the board’s ‘out’ to the usb-c connector’s d+ or d-.
|
|
#
?
Oct 11, 2021 07:19
|
|
|
|
| # ? Jun 11, 2024 09:53 |
|
|
 Seems like you connect CC1 and CC2 to ground each via 1 kohm resistors (which tells the host that there's an analog audio thingy there), you connect the mic's VCC to +5V, the mic's ground to ground, and the microphone output to SBU1. D+/D- are used as analog audio outputs and SBU2 is used as the return for it. You can ignore those three pins for just a microphone. I recommend reading it closely before you wire up anything because my actual experience building USB hardware is limited to USB 2.0 BattleMaster fucked around with this message at 07:47 on Oct 11, 2021 |
|
#
?
Oct 11, 2021 07:36
|
|