|
Just realized commercial insulating pads are made from silicone as well as mica- I have thin silicone sheet on hand and can easily cut a gasket of the stuff to suit. I think I'm comfortable attempting that and seeing how it performs before going with a commercial pad, the performance may suffer from it being thicker than is ideal but I can't see it being inferior to a commercial pad in terms of safety or risk.
|
# ? May 5, 2020 21:01 |
|
|
# ? Jun 12, 2024 20:48 |
|
If you have more than one lm338 you could split the load between them
|
# ? May 5, 2020 21:03 |
|
Ambrose Burnside posted:Just realized commercial insulating pads are made from silicone as well as mica- I have thin silicone sheet on hand and can easily cut a gasket of the stuff to suit. I think I'm comfortable attempting that and seeing how it performs before going with a commercial pad, the performance may suffer from it being thicker than is ideal but I can't see it being inferior to a commercial pad in terms of safety or risk. It doesn't hurt to try, so you should certainly do that, buuuuuut... "Silicone" actually encompasses a whole poo poo ton of different chemical formulas, there are a class of them specifically designed to have really high thermal conductivity, so your silicone probably will be inferior. But it might be Good Enough, anyway!
|
# ? May 5, 2020 21:39 |
|
ante posted:But it might be Good Enough, anyway! Yeah, that’s what i figure. As it stands an LM338 is overkill for this application, I almost went with an lm317 but the power reqs were cutting it close there and a good heatsink would be a lot more critical. I also have more 338s so if the pad doesn’t work out i’ll probably just take the easy path like taqueso suggested and chuck another regulator in there. i haven’t worked w components that need rigorous thermal management befpre so i kind of want to go through the motions here, where the stakes are low and things are relatively forgiving.
|
# ? May 5, 2020 21:59 |
|
Then do it again with a switching regulator and no heatsink
|
# ? May 5, 2020 22:01 |
Ambrose Burnside posted:Yeah, that’s what i figure. As it stands an LM338 is overkill for this application, I almost went with an lm317 but the power reqs were cutting it close there and a good heatsink would be a lot more critical. I also have more 338s so if the pad doesn’t work out i’ll probably just take the easy path like taqueso suggested and chuck another regulator in there. i haven’t worked w components that need rigorous thermal management befpre so i kind of want to go through the motions here, where the stakes are low and things are relatively forgiving. even w the pad i would still make sure you have everything well-enclosed even if it means adding a fan, you don't want a wire or a splash or whatever getting in the wide open "heatsink hole" even if the package casing isn't exposed as much anymore
|
|
# ? May 5, 2020 22:05 |
|
Does anyone know what could be causing bad pH readings using this setup? Taken from a TI guide. The pH probe produces a bipolar voltage (< 0 if acidic, > 0 if basic) The intent is the voltage ref + voltage divider on the right is to offset the voltage to make it unipolar, and the amplifiers are unity-gain, to deal with the high impedance, while leaving voltage unaffected. I'm having a hard trouble grokking it, but from what I gather, the point is so that you don't have to draw lots of current to deal with a high-impedance sensor. When I test individual parts of the circuit with a multimeter, everything appears to be working, as indicated in the image: - The vref + divider produces the right voltage - The pH probe when connected directly to the multimeter produces the right voltages - When the probe is replaced with a wire, the right voltage is read at the ADC (ie the .512v from the vref) Here's where it's weird: - I get ~2.5v to the ADC when the pH probe is replaced with nothing - I get a split second of a reasonable reading around .5v at the ADC when I plug the probe in sometimes, but as soon as I lightly touch the outside of the probe's wire, it shoots up to 1.5 to 2v and stays there. Or it just starts there. Hypothesis: There are still impedance issues causing whatever's generating the 2.5v open voltage (Not sure where that comes from, but must be from the top amplifier since the rest is disconnected) to override the probe voltage. Also of note, while measuring the voltage between the probe leads directly gives good readings, reading them between ground and the probe doesn't, which is presumably related to the need for unity gain amplifiers. Dominoes fucked around with this message at 00:19 on May 6, 2020 |
# ? May 5, 2020 23:52 |
|
The pH probe produces a differential output and you're trying to force one leg of its output to half supply voltage? Depending on how the probe works it may or may not be okay with that.
|
# ? May 6, 2020 00:22 |
|
This is how the TI article describes it, paraphrased : "The pH-measuring electrode produces a voltage that rides on top of this 512mV bias voltage." I've confirmed with the ADC datasheet that the inputs can't be negative, so something akin to this is required, ie offsetting with a voltage greater than the probe can produce.
|
# ? May 6, 2020 00:35 |
|
After thinking about your comment more and doing some more digging, I'm going to see if I can put this ADC's differential input capability to good use... edit: Got something almost-working (the pH values are a bit off, but the results are in the realm of reasonable) using a similar setup, but subbing out the vref for differential inputs. Dominoes fucked around with this message at 01:50 on May 6, 2020 |
# ? May 6, 2020 00:57 |
|
It looks like Eagle CAD is getting rolled into Fusion 360 and put on subscription-only pricing, which is bad in itself, but now one of my colleagues is telling me that when he bought a new computer and tried to move his old (pre AutoDesk) version of Eagle onto it they won't let him re-activate the software he already paid for. It feels like time to move on to another software package but it's been most of a decade since I've had to survey what's out there. I see KiCAD has gotten miles better and picked up CERN support which is cool, plus it's GPL licensed so it can't get taken away. DipTrace is out there too and reasonably priced. Are there any other PCB layout programs worth taking a look at as an Eagle-tier replacement? Altium/OrCAD are out of my range both in price and needed complexity.
|
# ? May 6, 2020 02:54 |
|
For your use-case, KiCad is good. For beginners, I'd usually recommend Upverter because it has many advantages, but it has the same syndrome as Eagle: cloud-based and recently picked up by industry leaders.
|
# ? May 6, 2020 03:03 |
|
PDP-1 posted:It looks like Eagle CAD is getting rolled into Fusion 360 and put on subscription-only pricing, which is bad in itself, but now one of my colleagues is telling me that when he bought a new computer and tried to move his old (pre AutoDesk) version of Eagle onto it they won't let him re-activate the software he already paid for. I use altium at work, but KiCAD is the only open source/free package worth learning, and my only hope is that one day it strikes down altium and pisses on it's grave
|
# ? May 6, 2020 10:19 |
|
Dominoes posted:After thinking about your comment more and doing some more digging, I'm going to see if I can put this ADC's differential input capability to good use... How are you calibrating your probe, 4/7/10 pH calibration fluid or against another "truth" sensor in a tank?
|
# ? May 6, 2020 15:26 |
|
csammis posted:How are you calibrating your probe, 4/7/10 pH calibration fluid or against another "truth" sensor in a tank? Still working on the calibration code / voltage-to-ph eq, but leaning known solns of arbitrary pH
|
# ? May 6, 2020 15:54 |
|
Dominoes posted:Still working on the calibration code / voltage-to-ph eq, but leaning known solns of arbitrary pH Use real calibration fluid and not an arbitrary pH source. Get some calibration fluid, soak your probe in a guaranteed known pH solution, read at the voltage coming off it, and only then model the bias. If you aren't using a known quantity (an actual known and trimmed value, not what another probe says in situ to a fluctuating environment) then you can't know that your voltage readings are "a bit off" because you aren't totally sure what the truth value is. From what I remember when I used pH probes on my Apex, every probe is going to be a bit different and they'll fall out of trim over time. You won't be able to determine the bias for one probe once and use it across the board. Be sure to consider this as you're building in a way for users to calibrate the probe.
|
# ? May 6, 2020 16:11 |
|
Are mylar film caps an acceptable substitute for ceramics in an RF application?
|
# ? May 6, 2020 17:26 |
|
RF stuff is pretty particular about cap construction. Without more details, let's go with "no"
|
# ? May 6, 2020 17:28 |
|
How about ceramics for films in an audio situation? It's those same 3 weird yet low tolerance caps in my recent post ITT from a late 1950s radio: 2500, 3000 and 5000 pF, all +-5%. This thing had surprisingly tight tolerances for the time. I'm guessing that they were film caps. Those 3 are axial and definitely ceramic caps in this radio are all radial. These 3 were around the volume pot. I had a tedious time hand grading caps to find ones in range of the old values. I came up with 2 ceramics and 1 film.
|
# ? May 6, 2020 17:37 |
|
The radio part, the audio part, or the power supply part? In order: no, yes, yes? I think you've already asked that ITT, so why don't you try it - No way to know without checking, but I doubt your ear would be able to pick up the difference
|
# ? May 6, 2020 17:48 |
|
Ambrose Burnside posted:Are mylar film caps an acceptable substitute for ceramics in an RF application? Like what ante said, it is probably safest to assume no, without details. Like if the capacitor appears in the 'RF signal path' of the circuit, and isn't just performing a DC blocking function, the RF circuit might be tuned to the particular value of 'parasitic' inductance of the capacitor, and changing the capacitor type to one with a very different parasitic inductance might change the circuit response in some way. I would say one aspect of detailed physical RF design which is kind of strange when compared to other kinds of circuit design is that parasitic aspects of wiring, circuit components, etc. often can't really be ignored or neglected and have to be absorbed into the circuit or tuned out by the circuit. silence_kit fucked around with this message at 18:09 on May 6, 2020 |
# ? May 6, 2020 18:02 |
|
I've seen some discussions of audio cap types and results, somewhere. Maybe I can did it up later. For audio you can certainly swap types and it won't be disastrous if power handling is ok.
|
# ? May 6, 2020 18:17 |
|
Ambrose Burnside posted:Are mylar film caps an acceptable substitute for ceramics in an RF application? If the RF application is some neat spark gap transmitter / vacuum tube tuned receiver / hand shaved crystal radio historical thingy like your other projects, maybe!
|
# ? May 6, 2020 18:48 |
|
Dominoes posted:Does anyone know what could be causing bad pH readings using this setup? Taken from a TI guide. Sorry I have no idea, but which probe are you using? I've been looking for something to help grow my basil and what not.
|
# ? May 6, 2020 19:59 |
|
csammis posted:Use real calibration fluid and not an arbitrary pH source. Get some calibration fluid, soak your probe in a guaranteed known pH solution, read at the voltage coming off it, and only then model the bias. If you aren't using a known quantity (an actual known and trimmed value, not what another probe says in situ to a fluctuating environment) then you can't know that your voltage readings are "a bit off" because you aren't totally sure what the truth value is. mobby_6kl posted:Sorry I have no idea, but which probe are you using? I've been looking for something to help grow my basil and what not. mobby_6kl posted:Sorry I have no idea, but which probe are you using? I've been looking for something to help grow my basil and what not. You'll find that a company called Atlas Scientific sells both probes and circuits on Amazon, but the circuit alone costs $40. Dominoes fucked around with this message at 21:57 on May 6, 2020 |
# ? May 6, 2020 21:53 |
Dominoes posted:Awesome. Bought some buffer soln off Amazon. From what I gather, calibration can keep it accurate over time, although response rate will decay. There's also a range of pH probes on the market using slightly diff approaches, and wildly diff prices, although they all seem to work by the same principle, and should be usable with the same electronics. I am not poo poo for electronics like you all but I do a lot of pH work in the lab etc. For Calomel electrode pH meters at least the solution / material temperature is part of the overall pH calculation. Does your probe also take this into account or if not does it need to? It's not a huge correction factor around room temp, but could be a concern if you were taking pH of some more extreme temps iirc. Also seconding using known standards for calibration.
|
|
# ? May 6, 2020 23:26 |
|
I'm taking temp into account. The main proj already has a temp probe, which makes this easy. I'm now designing a standalone module that incorporates an analog temp probe to go into one of the unused ADC inputs. From what I can tell, it doesn't seem to be a huge factor, so I figure a cheap probe with +-2°C tolerance will do. I think one of the reasons it's tough to find reliable info is that pH probes are used for very diff purposes. Ie the guides aimed at scientists make claims like "pH probes must be immersed in a xM KCl solution when not in use, and must be calibrated every use and replaced every so often", while ones targeted at growers etc may claim (what looks like a very similar probe with identical mechanism) can be submerged in the solution measuring indefinitely, and should be calibrated once a week. Dominoes fucked around with this message at 00:07 on May 7, 2020 |
# ? May 6, 2020 23:30 |
Ambrose Burnside posted:Are mylar film caps an acceptable substitute for ceramics in an RF application? Nah mylar is pretty garbage for RF. Unless you meant to say mica, in which case they're awesome.
|
|
# ? May 7, 2020 01:34 |
|
Dominoes posted:I'm taking temp into account. The main proj already has a temp probe, which makes this easy. I'm now designing a standalone module that incorporates an analog temp probe to go into one of the unused ADC inputs. From what I can tell, it doesn't seem to be a huge factor, so I figure a cheap probe with +-2°C tolerance will do. It's probably largely the same mechanics, but you're going from a lab specimen, +-.1pH (or better, I hope), to growers, where you're +-1pH. 7? Fine. 8? 6? Eh. probably still OK. 5? do something... Hobby-grade stuff really is "within an order of magnitude."
|
# ? May 7, 2020 01:35 |
|
I replaced the carbon brushes on my shopvac a few months ago. It wasn't a stunning success. The motor lacked it's full power and it sparked and smelled weird. Sometimes I would start at full power though, but loose it after a while. I finally opened it up again and there was carbon dust everywhere. I got some cheapo brushes from ebay and really they did not fit and where a PITA to replace because the motor looked like it was not designed to have the brushes replaced. I had to fit some spacers behind the springs of the new brushes so it would fit, I made them from brass. I am wondering if this might be the culprit behind the bad performance? Perhaps I should have tried to find copper, or aluminum even as they have better conductivity. I bought a new motor now and I will just replace it as I am tired of being without a properly working shopvac, but if I can fix this with some material replacement I would be keen on doing that, just so I can have a spare motor.
|
# ? May 7, 2020 06:04 |
|
Dominoes posted:From what I gather, calibration can keep it accurate over time, although response rate will decay. This is true as far as it goes but I can’t tell from your post if you’re taking into account the whole “recalibrate periodically” thing. Calibration is a way of generating biases and gains so that the output is accurate despite the physical reality that the input isn’t accurate, but any given calibration will only model the state of the probe at that moment. Since the nature of pH probes is that they change over time your initial calibration will become wrong over time. They aren’t just a set-it-and-forget-it device at the aquarium level of accuracy, where a misreading of 8 when the truth is 7.8 could be pretty bad for some organisms.
|
# ? May 7, 2020 13:50 |
|
Thank you for the details. I think I'll tackle this by experiment, including how often to calc. I'm considering using the Nernst eq constants (F, R) only for default calibration, and modelling voltage to pH as linear, with the slope and offset determined by 2 arbitrary (or constrained to match common buffers) calib pts. Would still need to include inverse Temp in the slope. From what I gather, linearity is generally gt 95%, but perhaps could use 3-pt nonlin calibration for more precision. Dominoes fucked around with this message at 19:22 on May 7, 2020 |
# ? May 7, 2020 19:14 |
|
ante posted:RF stuff is pretty particular about cap construction. i’m building baby’s first fm radio transmitter and am missing a couple of the specified caps for the preamp + inline with the aerial. i ended up substituting equivalent groups of smaller ceramic caps in parallel, it’s ugly but ehhh. had no trim caps for tuning so i’m gonna have to manually stretch/compress the air coil to change stations. if the mk1 works out and does something interesting i might get additionally stupid with it and modify the coil into a variable inductor by stretching/compressing the coil w a thumbscrew or wedges, but i’m getting ahead of myself. i’ve got an sdr dongle n accoutrement on hand so it’ll be extremely cool to properly observe the real-time effects of mucking with the tuning coil n so on
|
# ? May 7, 2020 20:30 |
|
That SDR for feedback on your circuit tuning idea is really good
|
# ? May 7, 2020 21:23 |
|
that’s the main appeal of the project to me- i’m living out in a rural area for a while and brought my SDR stuff expecting to take advantage of the lack of rf noise, but it’s actually worse than in a big city b/c there’s a radio tower nearby and i don’t have a band stop filter + my cabin is in a dell that seems to shelter me from line-pf-sight signals. so i’m gonna create my own rudimentary transmission source and experiment along those lines Ambrose Burnside fucked around with this message at 03:44 on May 8, 2020 |
# ? May 8, 2020 02:28 |
|
in other news, i’ve become hopelessly engrossed in making Pretty Component Subassemblies instead of just, building the fuckin circuit. weird things happen as i try to redirect my mental rudder from art metalworking to electronics that said i’ve figured out that this particular itch i have irt really baroque fabrication has a name, ”flywiring”, so i know i’m in good company
|
# ? May 8, 2020 03:48 |
Ambrose Burnside posted:in other news, i’ve become hopelessly engrossed in making Pretty Component Subassemblies instead of just, building the fuckin circuit. weird things happen as i try to redirect my mental rudder from art metalworking to electronics time to build some BEAM robots
|
|
# ? May 8, 2020 03:53 |
|
Nice
|
# ? May 8, 2020 03:55 |
|
Reminds me of the construction of this dude's discrete transistor pong. My reaction to this was "Holy poo poo this guy is German as gently caress". https://www.youtube.com/watch?v=Neky4fdaLhM
|
# ? May 8, 2020 04:40 |
|
|
# ? Jun 12, 2024 20:48 |
|
shovelbum posted:time to build some BEAM robots BEAM and dead bugging were my childhood electronics. I've still got a couple photovores around somewhere.
|
# ? May 8, 2020 14:56 |