@rawbrawb Today I though there was an exam for PLC class. We also have to design a modest PLC and submit in the exam day. I was up until 01:00 in the morning and then slept, woke up at 9:00 and studied until 11:00. Then I went to school, which takes 2 hours and 15 minutes!
And I was wrong, the exam is next week! Nobody was in the classroom!
Anyways, my project was not yet finished and I had no report at all. I am relieved now and writing the report for Butterworth HPF now, which is for my other course. Then I will make a PCB and test and submit on Wednesday.
I really need Oracal 651, transparent vinyl. That will make help me quickly create double-sided PCBs with ease! When it is transparent, aligning the layers is a piece of cake!
@yogece No video or material. Print using a laser printer, heat and press against freshly cleaned PCB copper. I use an HP LJ from a shop since my own printer is an inkjet.
@abdullahkahraman Open up one of the old type 7-segment LCD panels, look carefully at the glass in the area between the contacts at the edge and the digits themselves.
@abdullahkahraman LCD panels (like computer monitors) have wires that are so small they aren't noticeable. they also use ITO but it is a sheet (not patterned)
@abdullahkahraman One of the things about magician's tricks is the presentation. If the audience sees a largely transparent hand-held device, they'd forgive a small little blob of CoG ICs + components.
@rawbrawb any info regarding prototype.because my friend is doing his phd on some IC designing,he is using microwind,he is looking for the prototyping possibilities
@abdullahkahraman I have a different concern about transparent phones: Would the person on the other side of the phone be able to see everything that you read or type? Mirrored, of course.
@abdullahkahraman Though, an innovative solution would be to have two back to back displays: The one visible to others would show things like "On the phone" or "Do not disturb, boss on phone", and of course the current time. The one in front would be the regular phone display.
@yogece Just forget about using TSMC, they don't do business with you unless you can prove to them that you will run sufficient volume or your product has future volume upsides. MOSIS will probably be the least expensive for you,get into the shuttle program, and they will probably also provide libraries if you are doing digital.
@abdullahkahraman Do you actually believe the first transparent phones will be budget devices? If someone can make diamond encrusted cellphones, and they sell like mad, then saying "waste of resources" is a waste of resources
Who wanted to know about burning a bootloader using an Arduino?
@AnindoGhosh LAst week I saw a vendor who was promoting sapphire as the "glass" to use instead of gorilla glass. They claim to have a process to get the cost down.
@AnindoGhosh Of course no, they will be very expensive! I just think very selfish, if I don't like it, then it is a waste of resources, both in engineering time and material =)
@AnindoGhosh When Widlar was designing the first one, he used sheets of liquid crystal (mood ring material) to help him understand the thermal distribution and the thermal feedback for stability purposes. Interesting story for you...
@yogece I've never looked for any tutorials. There are whole courses available on small subsets of IC design and analog design like on coursera and other open learning initiatives from MIT for example.
@yogece you choose analog for some reason, which is the hardest area to start from, because of all the subtle aspects that have to be incorporated. It will take years before you can be sure of having something come out working. I'm not trying to discourage you, but it's not like picking up a programming language over the weekend. This takes concerted effort over years.
@rawbrawb so, another question i haven't asked on the site yet, but which puzzles me, is how this is a "shielded" antenna. It is indeed usually constructed with a coax shield, but exactly how is the "shielding" happening? It would seem to me that if you took any conductive loop, put a gap in it, then stripped the end of some coax, soldered the center conductor to one end of the gap and the shield to the other, you'd have an equivalent probe.
ie, the center conductor in half the loop isn't significant to the operation of the antenna; it's just a convenient construction method, to have the transmission line and the antenna be the same thing.
@PhilFrost so just to be clear, what you are describing could also be built by stripping back the outer layer of a coax, and then looping the central "tail" in a loop and soldering to the outer braid?
Just before you solder that loop back to the braid, you can imagine that there is a very large (relative) capacitance from that looped central conductor to the outer shield. So a real opportunity for an E-field to interact with it.
And once you solder that sucker down, you're thinking that it is shorted together therefore there is no capacitance. right?
i can see that before the center conductor is soldered to the shield, it's basically a dipole, formed by the center conductor and the outer surface of the shield
yeah, that makes sense, but I have a hard time intuitively understanding reception still, while transmission is easier.
so i guess any good conductor will try to gather all the E-field lines around it -- because it's a good conductor, the lines will tend to follow the conductor.
to get an antennae to radiate you have to impose upon it a (for lack of a better word - again a thought experiment) "standing wave" which really just means that by the time you signal gets to the top of the antennae the base has a different value, it is not a true equipotentail any more.
along it's length that it. SO the energy tries to make up for the difference and pops out into free space to make up for it. That's why physically short antennas often have loading coils at the top, to get the phase relationship between the top and bottom to be different.
That is also why slots can radiate, if you have a race track happening and the signal is propagating CCW and arrives later than the signal that is trying to get around the slot CW then you will get this phase difference along the slot and the E-Field pops out to try to balance things up.
In the early days in the marconi wireless on the big steam ships, you see pictures of this huge wire mesh structure that runs along the length of the ship. and people called that the antenae?
Well they were wrong, the antennae was the "feed wire" that ran vertically up to the "antennae" and the "antennae" was just a top loading capacitor.
In your loop, most of the central conductor really just deals with an internal e-field that is on the inside of the coax and the external e-fields don't get near it. Until you get to the gap, and that gap is electrically very small for the wavelengths you are dealing with.
@PhilFrost In the loop thingy we were talking about at the start above? No not really because it would be possible to have the top arc of the loop at a different phase than the point that is soldered to the outer shield. and so that wavelengths that you'd start to receive would be ~ 2X the diameter.
the claim that is that somehow an antenna like this rejects noise by being sensitive to the B-field but not the E-field. It seems like if you make it big enough to be resonant, then it's not really any different than a 1/2 wave dipole
so really it's a poor radiator, and the signal level will be less than with a resonant antenna, but as long as the signal is significantly more than the thermal noise floor, we can amplify it, so it works fine for receiving.
@PhilFrost but what is exposed to the E-Field is just that small gap. That is what the interaction is with. the fact that you have a loop figures into it in that it changes the "loading" impedance on either side of the gap, but it's on the same order of magnitude (vigourous arm waving going on here) as the size of the gap.
i don't know if i've ever seen anyone say that, but that's how they are always made. I think it's just because it's convenient, if you are making one from coax.
given that the inside and the outside of the shield are separate conductors at RF, I can't see how the center conductor inside the shield can be part of the antenna -- like you say, the gap is the antenna, and the loop serves to make a strong E-field right there that's coupled to the B-field in a strong way
@PhilFrost I wonder if you could just open up the central conductor for say 3 mm and then loop it around and solder that at the "base" of the loop?
@PhilFrost the only thing I can think of is that by having it on the opposite side, you are balancing the impedance on both sides so you're not getting that "race track" effect that is necessary to support the E-Field across the gap.
@PhilFrost I am rusty at this, I most care about this stuff for EMI/RFI reasons so fancy super tuned antenna theory is buries deeply. This last is getting out out of my depth.
so one thing I observed about the design illustrated in my question is this: imagine that you apply a voltage step to the feedline at the transmitter. That wave goes down the line and gets to the gap. There's a positive step on the center conductor, and the opposite negative step on the shield wraps back around the outside of the shield. They both travel around the loop, and cancel at the bottom where they meet.
@PhilFrost so at the red arrow, you are shorting the inner and outer conductors and then slobbering the whole thing to the feed line outer conductor right?
I'm not really sure because I haven't found much description of this probe beyond that picture, but I think what the inner conductor does is not relevant at ll.
@PhilFrost well ... if it is shorted you'll get a -'ve pulse reflected back and if it's open you'll get a positive pulse back. In either case it's a pretty "tuned" circuit.
@PhilFrost I hope I helped with my arm waving etc. I've got to go now.
I am sending it from 20mhz MCU with SPI freq being F_CPU/2, so each SPI clock should be 100ns in width. After the application of max propagation delay, worst case, the clock width is now 50ns. The receiving device is a TLC5940, which accepts data at up to 30mhz, meaning clock pulse width should be good to 30ns, i.e. <50ns, so it should all work. But it doesn't. Perhaps I have other problems?
@jippie Yes, exactly, but that 50ns half period can be eaten up by 30ns+20ns propagation delay.
@jippie I am not sure what the clock H required hold time is for TLC5940, all it says is that 30mzh clock max, it doesn't say what the clock H or L hold time is.
We have three phase input and the voltages in the three input is not the same always. Sometimes, one or two or three phase will be off. I would like to connect the phase to output which lies within the range (110 - 290).
Below will be the example,
Phase 1 (290v), Phase 2 (245v), Phase 3 (190v)
...
I've been having an issue with this SPI chip (MCP23S17 for days now. I am trying to interface it with a FPGA so my VHDL could be the root cause but I want to make sure my setup is correct
We have three phase input and the voltages in the three input is not the same always. Sometimes, one or two or three phase will be off. I would like to connect the phase to output which lies within the range (110 - 290).
Below will be the example,
Phase 1 (290v), Phase 2 (245v), Phase 3 (190v)
...
