Anybody know of a inline usb protection cable from overcurrent/voltage - Already had a previous laptop get destroyed and don't want it to happen again. I saw this topic electronics.stackexchange.com/questions/64321/… but I am wondering if there are already pre made solutions
Give it a shot. It's a longshot, but try it. You really shouldn't need the external Xtal: the factory chips come with fuses set to internal oscillator, but not holding the reset line high until you program might be a problem. So try it with the res.
However, given the variety of targets you have tried, the more likely problem (IMHO) is with your programmer, the NG.
@cpf BTW, you saw this right " On the Arduino Uno, you'll need to connect a 10 uF capacitor between reset and ground (after uploading the ArduinoISP sketch)"?
@cpf For when you are using the Uno as the programmer.
@cpf Right. Wait, so you can't program the NG, according to that website, because you can't get to NG's reset pin. So in theory, you should only be able to program the Uno from the NG, or the bare bones chips from either.
@cpf OK, so the cap needs to be in place after you've programmed the Uno with the ArduinoISP sketch: you put that in right before you use the Uno as the programmer.
I've an atmega328 chip (without bootloader), and I want to burn it with my Arduino UNO as ISP, so I can have a spare atmega 328 with UNO bootloader.
I'm trying to follow the instruction in http://arduino.cc/en/Tutorial/ArduinoISP, but there is a note that said:
Currently, you cannot use an...
Take the two (or three) 330ohm ones, and use them in parallel.
Three is better, but if all you got is two, then that will do.
330 is orange orange brown.
I don't know exactly how the reset line is wired up on the Uno, so I don't know exactly what kind of current we can pull from that, and how stiff it needs to be, so I am shooting for something close to what was specified in the post.
It's useful (I think) to get your entire set up actually set up, and then touch the connection andpoints continuity probes: this makes sure all the wires and the breadboard connections are good.
So I'd make sure that ATmega pins are connected to ATmega pins (UNO to barebones).
@cpf I have to run off soon. Maybe you want to add the whole capacitor/resistor thing to your question so that everyone knows?
@ThePhoton With two parts so similar, good usability dictates that the datasheet writers specify the application distinctions right up front when they are extolling the virtues of the devices.
@AnindoGhosh The first way I tried to answer your question was to go to one of the product web pages and look for a table of "similar products" ... no dice.
@AnindoGhosh Also, remember this part might be old enough that when the ds was written they assumed most readers would have the two datasheets back to back in a data book.
Also, the clear behavior is described in the 2nd paragraph on page 1 of each ds.
Maybe different designers have different philosophy about how restets should be used --- some just want to put the communication buffers in a known state, others want to reset everything, including the outputs under control.
@ThePhoton Actually neither of those is a TI original design, from what I can make out. There are equivalent historical parts, 595 and 596, from other manufacturers. This leads me to think there's something critical I'm not getting about the distinction.
@ThePhoton The TI innovation is the TPIC6A595 etc. Look at the current rating on the 6A*** beasties!
The TPIC6A595 can be used pretty much as-is as a motor driver, it even has high shunting clamps at 50 Volts, so no snubber needed. Amazing to have logic+power integrated like that.
@rawbrawb Interestingly, Tourette's Syndrome is commonly abbreviated as TS. An epiphany!
In the app note I just linked: "The TPIC6C596 is an enhancement of the TPIC6C595, where the shift register serial output (SER OUT) is clocked on the falling edge of the serial clock to provide additional hold time in applications where several devices are cascaded"
@ThePhoton Yes, and that is why I'm having to think about it.
I mean, I could of course just use bunches of H bridge ICs. However, being able to serial-in-parallel-out with 8 bits at power drive current levels is very attractive.
@MLM, it might be a bit over the top for what you need but I bought one of these a while back for debugging a system where the JTAG ground (and hence programmer) was at mains levels au.element14.com/analog-devices/eval-adum4160ebz/…
Probably not too hard (relatively) but an ATI GPU graphics card will give lower cost / hassle and faster speeds per $ from all I've seen. That sort of massively parallel integer is well suited to a GPU.
@PeterJ I'm never too worried about downvotes - there's at least a couple of people who at one time used to downvote everything I wrote, on principle, because of some differences of viewpoint around "real engineering" (theirs) versus "solve the problem" (mine) approaches.
First: Yes, your understanding is essentially correct, other than the issue being voltage and not charge.
Here is my analogy:
Consider a door to a house, with really smooth hinges, and no bolt or latch. The door is so light and so well-hinged that the slightest breeze would cause it to flap ope...
There we go.
"Real engineers consider it below their dignity to answer silly hobbyist questions, I'll just vote to close instead"... Supercilious ba-*-ds ;-)
@Kortuk Speaking of which... there was a recent question (not sure which one) where our bearded friend commented about "just using shields and stuff, not real engineering". I was planning to flag it, then I got distracted and lost the question.
@AnindoGhosh I can say for sure it was chance, on each post it had to have been different people. If you knew the days I could do a more detailed look for voting rings and such.
@AnindoGhosh because I know for sure that no single user did it. :) When you mentioned getting every post downvoted I verified no one did that to you. I have some secret tools :)
@AnindoGhosh nope, many real voting rings found happen to be a well hidden user with many accounts. We can find them, but they are generally not as obvious.
@AnindoGhosh I just dont like the though of voting rings. If you do ever seem to be revenge downvoted again let me know.
@AnindoGhosh I use ring to be more broad, it could be multiple people or just one. Generally multiple people is harder to find, but we still find them all with time.
@AnindoGhosh I have gone through 2 weeks of comments without finding it, it probably was already nuked.
@Kortuk If it ever happens now, I'd (a) be more comfortable passing over it and (b) flag it to you. I'm now aware that real people exist on chat, and also I'm that much less bothered by vindictive grouches ;-)
@Kortuk It was certainly not weeks - it was on one of my open tabs, and I only got back home and started opening tabs on the 2nd April. Then early this morning I did a windows update, so no more tabs.
@AnindoGhosh Ha. downvotes should generally be ignored as part of the noise generator on all communication media, but if someone is injecting a DC offset that is targeted at a single person I am going to clean it up to protect the speakers.
@AnindoGhosh Yeah, I did not realize for a moment how long I had made it back, then realized I had gone through two weeks and stopped looking
@Kortuk Then it's been nuked, which is a good thing - I remember thinking "This is the kind of comment that pisses duinoobs off". I wanted to tell you, but shiny things lured me off.
@Kortuk Wait, did you check closed questions? Because it was a question with 2 or 3 close votes when I opened it. Something definitely containing Arduino and Shield in the question, and shield(s) in the comment.
@AnindoGhosh you are still taking downvotes far too seriously. They are intentionally there to let people know their questions does not show research. That is clearly an example. Before I started posting on the site I took part for quite a while to watch and then took part when I felt like I understood what was going on.
@AnindoGhosh you have to admit, that is a terrible question.
+1 Because it's Friday, and I wanted to compensate for the unexplained -1. — David Kessner2 mins ago
@AnindoGhosh The point of the site is not to tell everyone, "Post your crap," We do want people to feel welcome, but they should not feel like they can just sling crap at us.
@AnindoGhosh it got what looks like a nice answer, I would consider that more important. I can see that you would like someone to explain why they downvoted.
@Kortuk While I'm all for getting people to improve post quality, I deal with newbies in several fields all the time, so I'm much more inclined to educate than to push one button on a power trip. Sorry, I think we'll always disagree on that.
@AnindoGhosh yeah, the best way is to edit and improve for them and inform them to try to do more like what example you have made. However, SE as a whole considers just accepting all crap to be a bad decision.
@AnindoGhosh just because someone has asked it before does not make it a well research questions. I can list hundreds of questions people have asked me to avoid doing any research.
Yep, that is separate to this. You can find a 555 timer circuit to get you a 1 hz blink pretty easily. Not saying it is instant. But if you know how to get it to blink you are only a step away. Just sharing the schematic of what he had would allow us to easily say, here is the one last bit you were needing.
@AnindoGhosh :( why such a sudden dropping it? I am just trying to discuss.
If the ATmega's fuses are set for using external oscillator, how would one reprogram it without using an oscillator, without resetting those fuse bits?
@DavidKessner That's a substitute oscillator... I'm sure the question is about dealing with a situation where one does not have an oscillator available. All clocks are oscillators, ideally - just not crystal oscillators. :-)
@AnindoGhosh Yea, so you need a substitute osc. There is almost always one available, since one is needed by whatever device is trying to do the programming.
I think that guy wants to reprogram it using the same universal programmer.
@AnindoGhosh It is ambiguous with the current Q. Basically, if the OP wants to reprogram with the universal programmer then the answer is correct, just badly worded.
@DavidKessner Here's my concern with that: An AVR MCU typically can be programmed without ext-osc being fed, by many different programmers. However, to my understanding, once one sets the fuses for ext-osc, then the MCU no longer works without such input, for programming or otherwise. Is my understanding wrong?
Basically, the MCU factory settings default to using the internal RC oscillator. Setting fuses for external oscillator disables that. Correct?
@AnindoGhosh I do not know about the AVR specifically, but every other MCU that I have used would allow for reprogramming without an external OSC even if the ext. osc bit is set.
I have some questions these are i am not sure exactly:
I have designed a board that has classical 2 side PCB design. Frequency is not a big issue for me but with ESD my CPU resets itself. (CPU clock 20 Mhz, and some data transfer lines at 10 Mhz, 2Mhz, 400 Khz )
1- How is returning current work...
That's a good Q, but for some reason I just don't feel like spending lots of time answering it when the guy uses the default Anonymous username. So much for the "Feeling good, it's a Friday" thing! :)
@DavidKessner Yup - once fuse bits are set for external oscillator, one needs a high voltage programmer to program the MCU if external clock is not available.
Sounds like the AVR has some sort of internal charge-pump that requires a clock to generate the higher voltages for programming the flash. And that a "High Voltage Programmer" provides the required voltage.
@DavidKessner The default programming mechanism used with AVRs uses the same voltages as the MCU's specified Vcc, and does not feed a clock - it basically does a serial programming (SPI essentially), not a clocked JTAG method.
well, it feeds a clock, but a serial clock, not the MCU clock.
@DavidKessner Oh good. Then the answer is merely ill-worded, rather than incorrect. The correct answer would be to the effect that the OP needs to override fusebits using the HV programming mode, then reprogram. However, since I don't know the HV stuff well (@jippie does I think) I won't jump in there.
@PeterJ No, I don't. The OP actually put that stuff in an answer (which might have gotten deleted by now). For whatever reason the OP didn't know or couldn't edit the Q directly.
@DavidKessner False. Insufficient facial hair to qualify.
@PeterJ 10 ms double flash is visible, and is actually more noticeable than a single 20 ms blink. I use that kind of double-flash for status indication on some designs - saves power while being difficult to ignore.
@DavidKessner verified - 10 ms on, 50 ms off, 10 ms on, 1 second off is my standard flash routine. It works. For distinguising the two flashes well, 75 ms gap between is enough.
@DavidKessner Yes, that's actually more noticeable to the eye than a straight 20 ms flash, while taking (approx) the same power.
Something to do with the physiology of vision - the eye notices change of intensity more than light alone, and a double-strobe is noticed more than a single strobe.
My 10-50-10 was based off an academic paper related to vision I had read somewhere... However, I personally notice a double-strobe with 60 to 70 ms dark time better than one with just 50 ms. Over 70, the impact is steady. This is based on some experiments I just did.
@AnindoGhosh I can understand a 50 ms dark time (still seems short, but doable), but the OP wants 10 ms where the LED is dark between blinks. I don't think that our vision system can see that as two distinct blinks.
@DavidKessner It cannot. Even at 50 ms dark time, it's not as though we can consciously distinguish the two blinks (I asked my wife to count the blinks). But yes, the double flash seems brighter than a single 20 ms blink.
@DavidKessner Have any MCU board handy? It's a really neat thing to experiment with. When I use 5 flashes in a row with 50 mS dark times, the effect is inexplicably irritating. That's what I've been using for error condition indications.
@DavidKessner There was this software jippie had mentioned - allows scripting to be run from the computer straight through to some MCU board. I can't remember what software and which board, but I remember downloading and installing it, playing with it and then deleting it. Very useful for quick experiments like this. No flashing of MCU needed on each experiment.
@DavidKessner Or get an Arduino! :-D :-D :-D Speaking of which, why did I not do this test on the duino? would have taken me half the time setting it up.
@DavidKessner I'm an FPGA challenged person. I'm sure that can be cured the day I get hold of an inexpensive FPGA board to experiment with, but until then, I don't speak FPGAese
@DavidKessner Can you imagine how much mark-up the most basic FPGA experimenter boards sell for in India? 500% and up!
@AnindoGhosh I dont mind sending electronics, the worry I always have is export restrictions and such, I have no idea how to tell what I can send over.
@AnindoGhosh I have had accellerometers and such before that I know I could not legally export, but only from reading on sites the notice of the export restriction, the product itself had nothing to warn me.
@AnindoGhosh There are some things that are restricted from export from the US. They don't want us exporting things that could be made into weapons-- and that includes some sensors, processors, etc.
@DavidKessner BTW we have a Mumbai HackerSpace, and it isn't really doing well, due to lack of infrastructure and dedicated space. I am a member, but not very happy with it.
@DavidKessner Yes, those fall into the FTC munitions definition.
Not sure if it is FTC, hold on.
OK, it is the "U. S. Bureau of Industry and Security" that enforces the restrictions, and apparently exporting the graphics accelerator card that is running on my PC, is illegal too. Anything with more than 4 processing cores, anything which can be used in space, and so on.
So my Intel i7 CPU is illegal too. This is amusing.
@Kortuk @DavidKessner Here you go: The official document: U. S. Bureau of Industry and Security
I am beginning to suspect that... A bunch of the product types mentioned in the document are manufactured in China by the respective companies. So it's fine for the US to import them from China but not for them to then leave the US, though they can just as well get to India direct from China? Ostrich/sand/head.
Ooooh, this is brilliant reading. Any optoelectronics containing "One or more than one internal “laser” diode;"... That nixes the laser mouse I have.
Okay, @DavidKessner You can send me your freebie FPGAs. The restriction reads:
‘Field programmable logic devices’ having any of the following: a.7.a. A maximum number of digital input/outputs greater than 200; or a.7.b. A system gate count of greater than 230,000;
@AnindoGhosh ITAR is far far more silly than those examples. Because we are multi-national, multi-jurisdictional when we deal with certain customers, they can't tell us the spec that they need because to do so would violate ITAR, because that spec may leak back out of the US.
@rawbrawb I'm just running a mental inventory of how many commonplace things in my house are not export-permitted. It's educational. Also, a surprising number of those things are patented by non-US companies.
So bringing in a device that is made out side of the US is even hard even though it out performs the existing eqt. because they can't confirm if you do out perform.
And a LOT of devices now do out perform. There is a bill trying to be pushed through that loosens off the restrictions because for the most part, it restricts the US's competitiveness.
@rawbrawb I can understand that. At the simplest level, I cannot legally import most Tektronix scopes, but I can get their far east equivalents, or better.
I've an atmega328 chip (without bootloader), and I want to burn it with my Arduino UNO as ISP, so I can have a spare atmega 328 with UNO bootloader.
I'm trying to follow the instruction in http://arduino.cc/en/Tutorial/ArduinoISP, but there is a note that said:
Currently, you cannot use an...
I know I can let an LED flash using a 555 timer, but how can I let it flash twice every second?
Asad, No you do not need the external oscillator to reprogram the flash.
I have some questions these are i am not sure exactly:
I have designed a board that has classical 2 side PCB design. Frequency is not a big issue for me but with ESD my CPU resets itself. (CPU clock 20 Mhz, and some data transfer lines at 10 Mhz, 2Mhz, 400 Khz )
1- How is returning current work...
