Electrical Engineering V3 - Developers

Seems a better gig than I've ever had. People knew I knew a thing or two about the angry pixies, but they always treated me like an idiot. Like the time I re-wired a 220V microwave oven transformer to put out 5V, and my boss said he didn't like me "playing with such high voltages" (we worked with 220V all the time, sometimes higher, installing air conditioners and high-power heaters). But then this is the same guy who swore up and down to a service tech that there should be continuity between the primary and secondary winding of a transformer, so...

[QUOTE=pentium;52395726]So you get to spend 12 hours walking around with a FLIR and watching the lines sag?[/QUOTE] I've had the same gig at a large lanparty, except with a IR thermometer monitoring temps of junction boxes.

[IMG]http://i11.photobucket.com/albums/a166/ballsandy/New%20Bucket/IMG_8103.jpg[/IMG] The motor isn't even seized. :huh:

I've seen contactor coils that just burned up out of nowhere like that, too. Hell, even brand-new ones that weren't in service for more than 5 months.

I just discovered [B]Charles Lohr. [/B]I figured that you guys might like his stuff anyway. He is currently doing reverse engineering of the HTC Vive and running the [URL="http://libsurvive"]https://github.com/cnlohr/libsurvive[/URL] which is a lightweight library for the Vive. He does a lot of stuff with ESP8xxx, lots of LED stuff and what not. Here is one of his livestreams - They are quite long. But try ti jump a bit into it. [video=youtube;RExji5EtSzE]https://youtu.be/RExji5EtSzE[/video] But I'll recommend to see one of his many interesting videos!

I finally finished my first real project. Most of my electrical work is spent fixing things rather than designing them, but here is an exception. I've needed a high current bench supply for a while now so I decided to rip apart a power supply from a 1U HP server and give it voltage and current control. Here are the final shots: [url=http://imgur.com/DwuN3dD][img]http://i.imgur.com/DwuN3dDh.jpg[/img][/url] [url=http://imgur.com/eECS8YJ][img]http://i.imgur.com/eECS8YJh.jpg[/img][/url] [url=http://imgur.com/mQqcZXg][img]http://i.imgur.com/mQqcZXgh.jpg[/img][/url] [url=http://imgur.com/KJ8LYPw][img]http://i.imgur.com/KJ8LYPwh.jpg[/img][/url] And the full gallery [URL="http://imgur.com/a/ywNre"]http://imgur.com/a/ywNre[/URL] I'm very happy as to how it turned out. It's simple but effective and I like being able to repurpose things to make them useful again. Also this provides an extremely important piece of equipment I need to build my next project which is very ambitious and fills me with doubt whether or not I can do it but I am giving it a shot.

Happy Canada day! Got a new toy to play with. [t]http://i.imgur.com/51JqF0T.jpg[/t] (it's the RAC Canada Day contest from 0000-2400 UTC July 1st if you're a ham and are interested)

Meanwhile I'm just sitting here on channel 15 on the CB. :(

[QUOTE=papkee;52420174]Happy Canada day! Got a new toy to play with. [t]http://i.imgur.com/51JqF0T.jpg[/t] (it's the RAC Canada Day contest from 0000-2400 UTC July 1st if you're a ham and are interested)[/QUOTE] Lucky! I'm still stuck here on my BaoFeng 2m/70cm handheld and a really long whip antenna.

So after three months of banging my head on my custom LED driver at work I've finally made a breakthrough yesterday. Guys Buck-Boost is so overrated (considering my duty was 0.56) I should've done pure Boost months ago (with a duty of 0.25!) and now I present my constant current 260W LED Driver! With atleast a 95% efficiency. [video=youtube;rhGDrpqX9XA]http://www.youtube.com/watch?v=rhGDrpqX9XA[/video] It feels awesome to have made my first professional board. :dance:

I'm always slightly amazed when my boards work. I always expect to have messed up some small detail that I overlooked in the datasheet and be stuck with three boards that don't work. If there was a US fab with prices comparable to OSHPark that would be amazing. No waiting two weeks.

[QUOTE=papkee;52447127]I'm always slightly amazed when my boards work. I always expect to have messed up some small detail that I overlooked in the datasheet and be stuck with three boards that don't work. If there was a US fab with prices comparable to OSHPark that would be amazing. No waiting two weeks.[/QUOTE] Advanced Circuits is also pretty good but I just like OSHPark because ease of usability and prices (Literally more "professional" PCB fabs don't have the online drop-file zip submission/error checking that OSHPark does and its kinda sad).

Okay, EE question I might of already answered. I have a run of cat 5e cable that runs up to a mast on the roof. I have installed a PTZ camera up there. Originally I had it wired so one wire was delivering +12v and two others were grounds. With the exception of one wire the rest were handling video, audio and RS-485. Being aware of voltage dip as I installed the camera last night I verified that I was within the limits of my camera and was getting a good 11.74v at the roof under load (and 12v solid without load). The camera was installed and behaved fine for the remainder of the evening and when I played with it a little more this morning. Then this afternoon while we were seeing over 30c I played with the controls again and couldn't pan left or right.. Power cycling the camera didn't change anything. IT simply could not move anywhere but up or down. I went up on the roof and checked the voltages again an I was now playing at 10.22v (again, with 12v without a load) which was way too low so thinking for some reason the less than ideal wire guage was causing a larger drop all the sudden I tied the grounds together and doubled up the +12 line by putting the single unused wire into use. No change in the voltage dip. The only thing I can suspect is that the heat is adding an additional resistance on the wiring and thus during the day the voltage drops below the ideal level but I've never seen a case this extreme. Also I specifically know it is not a fault with the camera itself. I brought the camera down from the roof and plugged it into a 12v supply on my bench and I could get X and Y axis movement out of it immediately.

[QUOTE=pentium;52448039]Okay, EE question I might of already answered. I have a run of cat 5e cable that runs up to a mast on the roof. I have installed a PTZ camera up there. Originally I had it wired so one wire was delivering +12v and two others were grounds. With the exception of one wire the rest were handling video, audio and RS-485. Being aware of voltage dip as I installed the camera last night I verified that I was within the limits of my camera and was getting a good 11.74v at the roof under load (and 12v solid without load). The camera was installed and behaved fine for the remainder of the evening and when I played with it a little more this morning. Then this afternoon while we were seeing over 30c I played with the controls again and couldn't pan left or right.. Power cycling the camera didn't change anything. IT simply could not move anywhere but up or down. I went up on the roof and checked the voltages again an I was now playing at 10.22v (again, with 12v without a load) which was way too low so thinking for some reason the less than ideal wire guage was causing a larger drop all the sudden I tied the grounds together and doubled up the +12 line by putting the single unused wire into use. No change in the voltage dip. The only thing I can suspect is that the heat is adding an additional resistance on the wiring and thus during the day the voltage drops below the ideal level but I've never seen a case this extreme. Also I specifically know it is not a fault with the camera itself. I brought the camera down from the roof and plugged it into a 12v supply on my bench and I could get X and Y axis movement out of it immediately.[/QUOTE] Yea definitely beef up the wire gauge to power the camera, maybe add some capacitors on the camera end. Also add some heat sinks (maybe the RAM adhesive style) on the motor drivers used for the pan/tilt as MOSFETs become less efficient the more they heat up.

[QUOTE=pentium;52448039]Okay, EE question I might of already answered. I have a run of cat 5e cable that runs up to a mast on the roof. I have installed a PTZ camera up there. Originally I had it wired so one wire was delivering +12v and two others were grounds. With the exception of one wire the rest were handling video, audio and RS-485. Being aware of voltage dip as I installed the camera last night I verified that I was within the limits of my camera and was getting a good 11.74v at the roof under load (and 12v solid without load). The camera was installed and behaved fine for the remainder of the evening and when I played with it a little more this morning. Then this afternoon while we were seeing over 30c I played with the controls again and couldn't pan left or right.. Power cycling the camera didn't change anything. IT simply could not move anywhere but up or down. I went up on the roof and checked the voltages again an I was now playing at 10.22v (again, with 12v without a load) which was way too low so thinking for some reason the less than ideal wire guage was causing a larger drop all the sudden I tied the grounds together and doubled up the +12 line by putting the single unused wire into use. No change in the voltage dip. The only thing I can suspect is that the heat is adding an additional resistance on the wiring and thus during the day the voltage drops below the ideal level but I've never seen a case this extreme. Also I specifically know it is not a fault with the camera itself. I brought the camera down from the roof and plugged it into a 12v supply on my bench and I could get X and Y axis movement out of it immediately.[/QUOTE] Power of ethernet - which can be used with cat5e - is typically done at much higher voltage so you can draw less current for the same amount of power. About 57VDC at around 600mA (25-30W). Is it possible or practical for you to get a 48V power supply and a buck converter on the camera end? If wired correctly, you might be able to use a POE injector and POE power splitter to work with your setup, examples: [url]http://www.ebay.com/itm/321546057675[/url] [url]http://www.ebay.com/itm/272217462419[/url]

This isn't a PoE device. This is literally just an analog system running over cat 5 because it was cheap. If I was reckless I would give it a 14V feed and the drop should put it in spec at the pole at all temperatures. Also, now that it has gone below 25c the voltage at the pole is up to 11.02v and the PTZ functions work again.

[QUOTE=pentium;52449083]This isn't a PoE device. This is literally just an analog system running over cat 5 because it was cheap. If I was reckless I would give it a 14V feed and the drop should put it in spec at the pole at all temperatures. Also, now that it has gone below 25c the voltage at the pole is up to 11.02v and the PTZ functions work again.[/QUOTE] Didn't mean to imply that the camera was POE. I did try to kinda explain why POE works and suggest that you use the POE injector and splitter in case you needed more power. You would pass your serial and video over the non POE pairs. The POE equipment was just to get more power to the camera. The stuff I linked were active POE, one being a 48vdc power supply and the other being what I assume is a buck dc dc converter to bring it back down to 12v1a for the actual camera.

I don't have the space on the pole for that. The junction box I'm running out of is already packed.

Does it necessarily require shielding? If not, you could always just use thermostat wire. You can get anywhere from 2-10 18 ga. conductors fairly cheaply for large rolls of the stuff, and it should be readily available at most hardware stores.

I can't rewire the thing. I roughed all my lines to the roof four years ago before I started building around the cabling. Running a new line means a lot of work. [IMG]http://i11.photobucket.com/albums/a166/ballsandy/around%20the%20house/CGS_0232.jpg[/IMG] [url=http://i11.photobucket.com/albums/a166/ballsandy/around%20the%20house/CGS_0232.jpg]Direct[/url] [IMG]http://i11.photobucket.com/albums/a166/ballsandy/around%20the%20house/CGS_0233.jpg[/IMG] [url=http://i11.photobucket.com/albums/a166/ballsandy/around%20the%20house/CGS_0233.jpg]Direct[/url] [IMG]http://i11.photobucket.com/albums/a166/ballsandy/New%20Bucket/CGS_7641.jpg[/IMG] [url=http://i11.photobucket.com/albums/a166/ballsandy/New%20Bucket/CGS_7641.jpg]Direct[/url] [IMG]http://i11.photobucket.com/albums/a166/ballsandy/New%20Bucket/CGS_7640.jpg[/IMG] [url=http://i11.photobucket.com/albums/a166/ballsandy/New%20Bucket/CGS_7640.jpg]Direct[/url] (I'm using the images above as an experiment. If the images break, use the direct link below each one) Anyways it failed again this morning at almost 10AM so I did a reading and we are sitting at 11.33v and slowly dropping as things warm up. the mention that MOSFETS require more current as they get warmer is something that might actually be the problem. The camera is powered directly by the DVR. Inside it has a DC-DC supply. My guess is that as it warms up outside the current draw does as well and the poor supply in the DVR can't handle the load so it starts notching down the voltage until it isn't constantly trying to melt down. I can work around this by splicing the 12v line for the roof to an external DC supply. the DVR is already at capacity with its 12v 5A supply, so I can just plug in a second power brick for 12V 2A.

MOSFETs are voltage controlled not current controlled, unless you're talking about the dynamic gate current but that does not really increase with temperature, did you measure the 11.33V at the source or at the load ? Also those images don't work direct or otherwise.

All testing was done up at the pole which is about thirty feet from the DVR. Again, if you disconnect the camera/PTZ assembly the voltage goes right up to a solid 12.00v

Guys I'm trying to repair this solar led light [url]http://imgur.com/a/yKtPb[/url] Basically not working at all. I checked the solar element, it gives of a solid 2V, even in dim light, so that's not it. Now I am a bit stumped on where to begin. Draw the circuits? Check the elements?

[QUOTE=Killuah;52457669]Guys I'm trying to repair this solar led light [url]http://imgur.com/a/yKtPb[/url] Basically not working at all. I checked the solar element, it gives of a solid 2V, even in dim light, so that's not it. Now I am a bit stumped on where to begin. Draw the circuits? Check the elements?[/QUOTE] Yes if you could draw up a schematic from the PCB traces it would help loads, it looks simple enough, only a few components.

I fixed it. The fucking switch had one cable "soldered" to the bare plastic. [editline]11th July 2017[/editline] I bent a replacement part out of chicken wire

Never would have thought it would be impossible to find old track-ball mice that people are willing to give away in this day and age, but sure enough, as soon as I decide that I want to try and recycle old computer mice to give my servos encoder wheels...

[img]http://i.imgur.com/61i3n6K.png[/img] Thanks Xilinx (the correct switch is -h apparently)

about to start my second year of uni, gonna start digging into my actual EE classes: fundamentals, electronic materials (quantum and classical physics behind electronic stuff), and digital logic design. My goal is to intern/work either in robotics or power (like Tesla and stuff). What kind of hands-on experiences should I do to gain the knowledge needed? This semester I'll probably be working on the avionics team in my rocketry club, so I'll gain some electronic communication skills, but how can I get my foot in the door on stuff like computer vision?

[QUOTE=thefreemann;52473399]about to start my second year of uni, gonna start digging into my actual EE classes: fundamentals, electronic materials (quantum and classical physics behind electronic stuff), and digital logic design. My goal is to intern/work either in robotics or power (like Tesla and stuff). What kind of hands-on experiences should I do to gain the knowledge needed? This semester I'll probably be working on the avionics team in my rocketry club, so I'll gain some electronic communication skills, but how can I get my foot in the door on stuff like computer vision?[/QUOTE] Whew that's a large swathe of stuff, really depends upon where on the chain of robotics you want to be. Do you want to be down low level with all the nitty gritty electronics & analog stuff or high level machine intelligence/CV?

[QUOTE=LoneWolf_Recon;52473429]Whew that's a large swathe of stuff, really depends upon where on the chain of robotics you want to be. Do you want to be down low level with all the nitty gritty electronics & analog stuff or high level machine intelligence/CV?[/QUOTE] I'm really not sure at this point, not having any actual hands-on experience in either, I have an interest in both honestly, what would either route entail? I'm guessing the latter would require some fairly strong programming background?