|
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. Very old post but I don't know many other people with gan experience. Brining up my new design Transphorm's 650V Gan fets (in 8x8mm SMT packages) are destroying my original Infineon part choice which overheat at no load from switching losses alone (they do better with load as the topology starts ZVSing but no load must be supported). Anyways this has me looking more seriously at Gan in general and possibly for my lower V higher current secondary which is marginal thermally. I know you have experience with EPC, do you have experience or comments on Gan systems? Somehow they weren't on my radar. They make enhancement mode 100V and 650V devices and with somewhat higher gate voltage tolerance than EPC.
|
#
?
Aug 19, 2017 03:34
|
|
|
|
| # ? Jun 4, 2024 21:30 |
|
|
Tres Burritos posted:Scanned through the last couple of pages and the OP and didn't see this mentioned anywhere: There are cheaper 2 channel scopes, but the additional capabilities brought by the other 2 channels are well worth the cost difference. Martytoof posted:It does indeed. I have a base understanding of how to tap the power from the existing circuitry and what not to touch but I'm going to do all the actual work supervised. My father is an EE so I can get his input once it's time to crack everything open. A good plan. Running some quick numbers, at 5000v (typical ish), and 40watts (about as large as a consumer one come) you'd be talking about 8ma current being sourced.(halve to quarter that for the more common 10-20w ones) Unpleasant, but unlikely to injure. The power supply itself won't be built large enough to deal with the much higher load your body will subject it to, and will sag. It'll still probably give you a continuous shock, like an electrostatic generator. If it has a capacitor on the output (and the plates count as a capacitor), you'll get an instantaneous current much higher than that, which will be painful, and potentially injurious depending on just how much capacitance. All in all, I suspect the 120/240v input area is more dangerous once you've opened the case up, but caution is still advised.
|
|
#
?
Aug 19, 2017 04:49
|
|
|
Tres Burritos posted:Scanned through the last couple of pages and the OP and didn't see this mentioned anywhere: Yeah it's fantastic. If you're used to debugging circuits with just a multimeter it's life-changing. Get one. http://gotroot.ca/rigol/riglol/
|
|
#
?
Aug 19, 2017 18:54
|
|
|
Would a signal generator be a silly thing to buy for helping to learn some basics? I figure a signal generator + make a little amplifier would be a fun / good project for learning on an oscilloscope right? Any recommendations on that front?
|
|
#
?
Aug 20, 2017 20:41
|
|
|
We used a function generator quite a bit in my first couple EE labs, its a decent way to characterize a circuit empirically. I don't have a suggestion on an exact model, but something from Rigol will probably have a good feature/cost ratio.
|
|
#
?
Aug 20, 2017 20:45
|
|
|
A function generator probably isn't that useful to buy at first. Especially if you're looking to build something, and have a scope. It's fun and good practice to build your own out of a 555  And maybe some filters to get some neat waveforms.
|
|
#
?
Aug 20, 2017 22:24
|
|
|
ante posted:A function generator probably isn't that useful to buy at first. Especially if you're looking to build something, and have a scope. This. A function generator is overkill. You're not trying to calibrate your scope with a known input, and any signal you need to see, you should already be generating yourself.
|
|
#
?
Aug 21, 2017 05:36
|
|
|
Aurium posted:There are cheaper 2 channel scopes, but the additional capabilities brought by the other 2 channels are well worth the cost difference.
|
|
#
?
Aug 21, 2017 15:35
|
|
|
KnifeWrench posted:This. A function generator is overkill. You're not trying to calibrate your scope with a known input, and any signal you need to see, you should already be generating yourself. Don't get a real function generator, get one of those DIY function generator kits (or build one from a 555 as was mentioned). They're sub-$50 and can do all the function generate-y things you'd actually want and need at this stage.
|
|
#
?
Aug 21, 2017 15:43
|
|
|
Sagebrush posted:Yeah it's fantastic. If you're used to debugging circuits with just a multimeter it's life-changing. Get one. Seconding this, I keep one at home. The value is fantastic.
|
|
#
?
Aug 22, 2017 19:55
|
|
|
I upgraded my heater controller today. Someone in the thread pointed out that the higher than expected temperatures were probably being caused by the ESP8266 getting warm next to the temperature sensor. So now there's a second ESP8266 with temperature sensor in the room, which communicates over the WIFI using UDP to give the heater controller a more accurate temperature (which is roughly two degrees lower, as expected). The original temperature sensor inside the enclosure now acts solely as a safety device or backup reading if the other sensor is not available. ESP8266s are fantastic, I was able to throw this together in just a few hours. Splode fucked around with this message at 02:41 on Aug 24, 2017 |
|
#
?
Aug 24, 2017 02:38
|
|
|
I'm planning to solve the issue of my main hallway having no natural light going through it by setting up two panels of daylight led strips behind frosted glass on the ceiling to look sort of like skylights (and hopefully not like industrial office lights). While I've seen a few instructions for making panels similar to these, they have tended to be for photography/video purposes, and not for recreating actual daylight. Has anyone here done anything like this and is in a position to comment on my plans? While i know its important to get the right light level (6500k for daylight LEDs), some extra research suggested the CRI levels to be important to get it to look like daylight. While a specific brand had as high as 97, the LED strips I've been looking at only suggest 80+, and i don't know if that sufficient for my purposes. So I'm thinking that it would run from the mains in my hallway, up go a control switch on the wall like a light switch, and up onto two panels in the ceiling (guessing at two 50cm by 50cm panels). On a basic level i would like on/off and dimming capabilities, but I'm not sure whether PWM or voltage regulation would be the better approach for this, and words would be welcome? If i got that to work, i might even toy with getting WiFi/Arduino involved, to allow control via mobile phone, or another light setting which sees an automatic rise and drop in brightness to match the daylight hours at the day. It feels a bit gimmicky, but if i was in all day, a natural rise and fall of light without me having to adjust it might be nice. Has anyone got some experience of doing this, with a focus on the mimicking of daylight. I appreciate the quality of the light will probably come down to the quality of the LEDs (high CRI LEDs cost a lot more than normal ones, and I'd want close to 10m worth), but any other tips or advice? While I've seen lots of things about building panels, not match the particular aims I've mentioned here.
|
|
#
?
Aug 30, 2017 16:41
|
|
|
So I'm fairly new to electronics. I've been modding Gameboys for the past couple months to have backlights. Here's an example of one:  (It looks better in person.) Now, my difficulty is in installing the bivert mod chip. This chip adds better contrast to the screen and makes it look loads better. The problem I'm running into is desoldering to lift a couple pins off the board. Here's what it the board/pins look like: https://youtu.be/xS8NQ3dds2k (mobile YouTube doesn't allow time-stamping. Skip to 5:15.) I have the desoldering braid and a cheap soldering iron that's worked for most everything else too. Do I need to get a hotter soldering iron? I'm thinking of getting a 60W.
|
|
#
?
Aug 30, 2017 16:51
|
|
|
Fat Turkey posted:I'm planning to solve the issue of my main hallway having no natural light going through it by setting up two panels of daylight led strips behind frosted glass on the ceiling to look sort of like skylights (and hopefully not like industrial office lights). While I've seen a few instructions for making panels similar to these, they have tended to be for photography/video purposes, and not for recreating actual daylight. Has anyone here done anything like this and is in a position to comment on my plans? I partially did one of these: https://www.youtube.com/watch?v=Y2KK4YiOO1o Made it out of an old TV (be careful, the LED drivers in larger TV's can deliver 100+ volts) and it works reasonably well I just never got around to shoving the bits back together and hanging it up. e: also this guy just solders the PWM brightness control to the positive rail which means it's locked at full brightness all the time, if you have an arduino you could actually feed a PWM signal into it and make it adjustable.
|
|
#
?
Aug 30, 2017 16:56
|
|
|
Detective No. 27 posted:So I'm fairly new to electronics. I've been modding Gameboys for the past couple months to have backlights. A better soldering iron will probably help. The Hakko FX888 is one of the most recommended for hobby level stuff, even though it's a little bit expensive. I got a cheaper knock off soldering station and it's been a little frustrating. There's also the new TS100 portable soldering iron that's been mentioned up thread. It's a little more DIY with the power delivery to it, but it sounds pretty nice. I'm probably going to pick one up sometime. Here's an amazon link but they're probably cheaper on ebay: https://smile.amazon.com/Programmable-Pocket-size-Soldering-Interface-Acceleration/dp/B01MDTO6X7/ To help flow the old solder you can heat it and add a little new solder to make it easier to melt, but you'll have to be careful about bridging to the other pins. Also, desoldering braid is great, especially on surface mount connections, but if you have issues with it the sold-a-pult style spring loaded suckers are pretty decent. You just get the solder liquid and hit the button and it pulls it off the board with a quick vacuum action.
|
|
#
?
Aug 30, 2017 18:29
|
|
|
Oh yeah, I've got a solder sucker. I'm gonna try one more thing tonight, and lift the pins with a dental pick. If that doesn't work, I'll start saving for that soldering iron. Quite pricey. I'll post a status update with the results.
|
|
#
?
Aug 30, 2017 19:22
|
|
|
Detective No. 27 posted:Oh yeah, I've got a solder sucker. I'm gonna try one more thing tonight, and lift the pins with a dental pick. If that doesn't work, I'll start saving for that soldering iron. Quite pricey. I'll post a status update with the results. Something to keep in mind when looking at the price of a Hakko (which I also recommend) or some other temperature controlled soldering station is that having the right temperature extends the life of your tip significantly. I used to go through several tips per year with a Radio Shack pencil iron doing a tiny fraction of the volume my Hakko has handled without a tip change. Wasted heat on an iron means increasing wear and tear on the tips, which add up over time.
|
|
#
?
Aug 30, 2017 20:12
|
|
|
https://www.youtube.com/watch?v=HgrB5P-rDLw
|
|
#
?
Aug 31, 2017 00:21
|
|
|
I got a TS100 and I genuinely like it however I still prefer my Hakko as a daily driver, just because it feels a bit sturdier and the tips are an order of magnitude cheaper so I could get a pack of like 10 lovely tips for $5 vs. $10-$20 each for the TS100
|
|
#
?
Aug 31, 2017 00:44
|
|
|
ate all the Oreos posted:I partially did one of these: https://www.youtube.com/watch?v=Y2KK4YiOO1o Haha yeah I've been watching that guys videos and it's partially an inspiration for this. I don't have access to old laptops or TVs and felt using LED strips of sufficient brightness would be the safer and easier method. I guess part of the learning curve is I'm not exactly sure how bright i want the LEDs to be. While i want it to simulate daylight in a dark hall, i don't want it to be like staring into the midday sun! But a frosted covering should spread the light sufficiently for that. Yeah i wasn't sure about his PWM approach as being efficient which is why i wondered about voltage regulation. Not quite sure how i could get the output of an Arduino be either more efficient and/or actually power the LEDs (they'll need a 12-24 volt supply, which Arduino doesn't output), so i guess i need to read up how that output could trigger the supply. I'm pretty certain I'll at least attempt the project and can document it for others.
|
|
#
?
Aug 31, 2017 03:05
|
|
|
Fat Turkey posted:Haha yeah I've been watching that guys videos and it's partially an inspiration for this. I don't have access to old laptops or TVs and felt using LED strips of sufficient brightness would be the safer and easier method. Voltage regulation doesn't work for LEDs. You'll just cross a voltage threshold point and they'll turn off. Their brightness is instead controlled using either current (in a fixed circuit) or PWM (as turning them on and off rapidly adequately tricks your brain). This is why people have trouble when they try and use old school light bulb dimming circuits to control LED ceiling lights. They just flicker weirdly rather than dim.
|
|
#
?
Aug 31, 2017 04:59
|
|
|
poeticoddity posted:Something to keep in mind when looking at the price of a Hakko (which I also recommend) or some other temperature controlled soldering station is that having the right temperature extends the life of your tip significantly. I used to go through several tips per year with a Radio Shack pencil iron doing a tiny fraction of the volume my Hakko has handled without a tip change. Wasted heat on an iron means increasing wear and tear on the tips, which add up over time. Put a blob of solder on the tip before you set it back into its holder, and it will last approximately forever. The blob will radiate heat and oxidize while still leaving nice shiny solder underneath that's actually in contact with the tip. When you're ready to make another connection, just tap the excess solder off into [some container], stab it into your bronze wool cleaner a couple of times, and have a perfectly tinned tip that's ready for action every single time.
|
|
#
?
Aug 31, 2017 15:56
|
|
|
Acid Reflux posted:Put a blob of solder on the tip before you set it back into its holder, and it will last approximately forever. The blob will radiate heat and oxidize while still leaving nice shiny solder underneath that's actually in contact with the tip. When you're ready to make another connection, just tap the excess solder off into [some container], stab it into your bronze wool cleaner a couple of times, and have a perfectly tinned tip that's ready for action every single time. Yeah soldering iron tips are made of, well, iron, because weirdly it's one of the only metals that doesn't dissolve into molten lead. This combined with getting hot means that they oxidize super fuckin' fast, but making sure you always leave them coated when you're done goes a long way to improve lifetimes
|
|
#
?
Aug 31, 2017 18:09
|
|
|
My experience echos that of poeticodditys - I think the cheap irons don't have good enough thermal regulation and/or a lovely coating, such that the tips will ruin themselves fairly quickly even if you baby them. I used to think I was doing something wrong and went through like 3 of them until I bought a Metcal - it's just one of those false economy things where buying cheap ends up costing more in the long run.
|
|
#
?
Aug 31, 2017 20:09
|
|
|
Acid Reflux posted:Put a blob of solder on the tip before you set it back into its holder, and it will last approximately forever. The blob will radiate heat and oxidize while still leaving nice shiny solder underneath that's actually in contact with the tip. When you're ready to make another connection, just tap the excess solder off into [some container], stab it into your bronze wool cleaner a couple of times, and have a perfectly tinned tip that's ready for action every single time. That helps, but running an iron significantly hotter than the minimum effective temperature for soldering increases the oxidation rate by orders of magnitude. My tip treatments were similar between my last pencil iron and my Hakko, but I burned through 4 tips with the pencil doing less soldering than my Hakko has seen on a tip that still looks almost new after several years. If someone's in the position where a temperature controlled system repesents a significant financial investment, it becomes a bit more palatable when they realize that they'll save on tip replacements in a relatively short time frame.
|
|
#
?
Aug 31, 2017 21:54
|
|
|
I maybe should have been more specific, that technique is really aimed toward maintaining soldering station tips that are constantly subjected to high temps. I don't think anyone should have to try to preserve the integrity of a non-regulated tip for very long if they can help it at all.
|
|
#
?
Aug 31, 2017 22:12
|
|
|
I am etching brass/bronze/nickel silver using a 30V 10A adjustable DC power supply but I can't seem to get the amps anywhere near what people talk about using when they etch. When I etch nickel silver, I etch at 15-20V and I get about .100-.200A. When I etch brass or bronze at the same voltage I get roughly .040A and then it quickly drops to 0 and stops etching. I can't for the life of my figure out why this happens, or how to get the amperage higher. I have the knobs for the power supply on full amperage. I am using salt water with hydrogen peroxide as the electrolyte which is what most people use. I connect the positive lead to the workpiece, and the negative one to the handpiece that has an electrolyte soaked rag on it, the same way they do with professional marking machines. Typically in tutorials, people talk about using anywhere from 2-24v with a 10A battery/car charger/power supply. PS. With 15-24V at lets say 1A, will I die?
|
|
#
?
Sep 4, 2017 23:11
|
|
|
Brekelefuw posted:I am etching brass/bronze/nickel silver using a 30V 10A adjustable DC power supply but I can't seem to get the amps anywhere near what people talk about using when they etch. I've never etched anything before but what happens when you move the hand piece closer to the workpiece? Also, what is your salt concentration versus what industry uses?
|
|
#
?
Sep 4, 2017 23:20
|
|
|
The handpiece is directly on the workpiece, which is covered with a vinyl stencil as a resist to get the pattern I want. My handpiece is more fancy than a q-tip, because when I used a q-tip, the current would erode the alligator clips because the salt water touched them directly. Now I have it so the alligator clip is far away from the actual liquid. https://www.youtube.com/watch?v=bV-PItEcLXY This is basically what I am doing, but with a power supply rather than a battery, and with copper alloys rather than steel. Nickel silver works great. I can get a .5mm etch in a few minutes, but the same thing on brass/bronze gets me almost no depth because the amps drop immediately.
|
|
#
?
Sep 4, 2017 23:33
|
|
|
Your connection to the metal is probably high resistance. Try scrubbing it with scotch brite or even just digging the alligator clips real hard. No, you won't be able to hurt yourself with 20V. Probably don't stick your tongue in there though.
|
|
#
?
Sep 5, 2017 00:05
|
|
|
If you manage to electrolyze salty water it will start generating chlorine gas, as I found out myself as a kid  Not a huge amount, just have a fan or something to blow it away
|
|
#
?
Sep 5, 2017 00:20
|
|
|
ante posted:Your connection to the metal is probably high resistance. Try scrubbing it with scotch brite or even just digging the alligator clips real hard. Would the thickness of the metal affect the resistance? The brass I tried to etch was .625" solid rod, and the Bronze was .030" sheet. Both etched poorly. The nickel silver that etches great is .020" sheet.
|
|
#
?
Sep 5, 2017 00:37
|
|
|
The massive hurricane that might be hitting me in the coming week has got me thinking, would it be "safe" to series up some 18650's to 12V and then run a power inverter off it? I have some actually nice 18650's made by reputable manufacturers that I trust not to simply explode on me, but if I were to draw, say, 100W through them would they catch fire anyway due to series resistance?
|
|
#
?
Sep 5, 2017 17:20
|
|
|
ate all the Oreos posted:The massive hurricane that might be hitting me in the coming week has got me thinking, would it be "safe" to series up some 18650's to 12V and then run a power inverter off it? I have some actually nice 18650's made by reputable manufacturers that I trust not to simply explode on me, but if I were to draw, say, 100W through them would they catch fire anyway due to series resistance? Series resistance won't do much to 18650s. How many are you running in parallel? Modern high quality 18650s like Panasonic can each draw 10 amps no problem, so technically with at least 2p you'll be fine to draw 100w. It gets tricky because a typical car battery inverter is expecting the battery to be between something like 12-14.5v, and shuts down around 11.5v. a low 18650 is 3.6/3.7 nominal but can safely be between 3.4 and 4.2 depending on state of charge, so three in series might shut down the inverter prematurely due to undervoltage, while four fully charged might not start the inverter due to overvoltage. So you might need to add some sort of regulator to that.
|
|
#
?
Sep 5, 2017 18:00
|
|
|
Zero VGS posted:Series resistance won't do much to 18650s. How many are you running in parallel? I was just throwing the idea around in my head so I'm not actually experimenting with it yet, but I think I have enough "good" batteries to do at least two parallel sets of 4 in series. I'll have to see where the cutoffs are for my inverter, I didn't realize the window was so narrow (since I figured car voltages are pretty crazy and float all over the place anyway). I'll report back if I burn down my apartment
|
|
#
?
Sep 5, 2017 18:19
|
|
|
Brekelefuw posted:Would the thickness of the metal affect the resistance? I don't have any domain knowledget that would help, but this sounds cool and I want to hear how it goes. Yes, thickness affects resistance. Resistance will be based on the material and the cross-sectional area of the "wire" between the conductors (even if the "wire" is a plate with a gator clip on it). Based on what you said, my uninformed guess is that some part of your hookup is corroding before you get the etching you want. I might try an etch, wait for the current to drop, then disconnect power and do resistance measurements around each joint in the system. Maybe the gator clip on the other side of the part is getting fouled up somehow?
|
|
#
?
Sep 5, 2017 19:34
|
|
|
ate all the Oreos posted:I was just throwing the idea around in my head so I'm not actually experimenting with it yet, but I think I have enough "good" batteries to do at least two parallel sets of 4 in series. I'll have to see where the cutoffs are for my inverter, I didn't realize the window was so narrow (since I figured car voltages are pretty crazy and float all over the place anyway). I'll report back if I burn down my apartment Yeah, I just googled a high-end 12v inverter and it has the cutoff range 10v up to 15v, so that might work OK with 3 in series, it'd cut off once the voltage sags below 3.33 per cell which is about where they're dead anyway. What are you trying to power? I run my soldering iron off this QD188 power bank and it is loving amazing, adjustable output from 5-24v and you supply up to 8 of your own 18650 cells to power it: https://www.aliexpress.com/store/pr...2485864933.html Just make sure your cells are matched to each other and it should work well. It can power a laptop for over 24 hours or a phone for over a week.
|
|
#
?
Sep 5, 2017 19:51
|
|
|
I've seen several manufacturers use P-clamps to clamp coax shields to chassis in the middle of the cable, kind of like this: Anyone know what kind of tool you'd need to strip the insulation cleanly like that, and what they're called?
|
|
#
?
Sep 5, 2017 21:23
|
|
|
Sometimes coax strippers, sometimes a knife. Don't over think it, just use box cutters
|
|
#
?
Sep 5, 2017 21:26
|
|
|
|
| # ? Jun 4, 2024 21:30 |
|
|
Stabby McDamage posted:I don't have any domain knowledget that would help, but this sounds cool and I want to hear how it goes.  Here is my setup for a successful etch on nickel silver. No surface prep needed. The positive lead goes anywhere on the part and the negative is attached to the aluminium handpiece with the salt water soaked cloth on it. I hold it over the vinyl stencil and check every few seconds for depth and quality. More volts is a more aggressive etch and not as nice looking as less volts. The etches I was trying this weekend were just on scrap brass, bronze and nickel silver. The nickel silver is the only metal that has a great looking etch. The rest etch quickly and then the amps go to 0. I think it's forming an oxide layer or something, which is preventing conductivity.
|
|
#
?
Sep 5, 2017 21:45
|
|