Back
[02:39:13] <STS_Patrik> whats the suggested method of measuring vibrations using a accelerometer (mpu-9150 in this case) ?
[02:41:56] <apo> STS_Patrik: You want to determine if it's vibrating at all?
[02:42:06] <apo> STS_Patrik: Try getting some samples and running a fourier transform
[02:42:45] <STS_Patrik> i need to get a value that can be compared between different machines
[02:42:49] <jacekowski> STS_Patrik: there are dedicated vibration meters
[02:43:03] <jacekowski> STS_Patrik: accelerometer might not be good enough
[02:43:48] <STS_Patrik> my hardware already have a different purpose on those machines, and it has a accelerometer already so it would just be a bonus to be able to get an indication of vibrations as well
[02:43:56] <jacekowski> but, if it works, it's going to be a matter of recording total acceleration and analysing it
[02:44:17] <jacekowski> STS_Patrik: sample as fast as you can, and then FFT it
[02:44:21] <jacekowski> STS_Patrik: and see if it works
[06:16:44] <LeoNerd> Mmm.. more boards shipping in my direction now :) Though I doubt they'll arrive before 2015
[06:17:50] <LeoNerd> Oh.. one random thing I noticed with my HVSP board: it seems an ATtiny chip itself has a reistance between Vcc and RESET - probably internal pullup?
[06:18:05] <Lambda_Aurigae> probably a weak one.
[06:18:12] <LeoNerd> If put e.g. an ATtiny84 in the socket and give it Vtg without Vprg, the LED I have on the Vprg line still justabout glows weakly
[06:18:34] <LeoNerd> I thought it was a mistake with my 2-transistor control circuit for it, until I notice it goes away if I remove the chip
[06:20:04] <Lambda_Aurigae> page 66 of the datasheet.
[06:20:13] <LeoNerd> I wonder if I should move the LED to a different part of the circuit then
[06:20:14] <Lambda_Aurigae> RTFM
[06:20:16] <Lambda_Aurigae> it's in there.
[06:20:44] <Lambda_Aurigae> RESET
[06:20:45] <Lambda_Aurigae> : External Reset
[06:20:45] <Lambda_Aurigae> input is active low and enabled by unprogramming (“1”) the
[06:20:45] <Lambda_Aurigae> RSTDISBL Fuse. Pullup is activated and output driver and digital input are deactivated when
[06:20:45] <Lambda_Aurigae> the pin is used as the RESET
[06:20:45] <Lambda_Aurigae> pin.
[06:21:41] <LeoNerd> Oooh.. right; yes. The fuse
[06:21:57] <LeoNerd> Ah that'd explain it - usually I work with my specially-labeled tiny84 with that fuse turned off
[06:24:44] <LeoNerd> Hmm.. so maybe I should move the status LED to higher up in the control logic, so as to avoid that..
[06:25:07] <Lambda_Aurigae> or, it can tell you that your chip has the fuse unprogrammed when you turn it on.
[06:25:53] <LeoNerd> Mmm.. subtle
[06:26:04] <Lambda_Aurigae> it's not a design flaw...it's a feature!
[06:26:06] <LeoNerd> Though I was thinking of having a controller board to do that... a few buttons and LEDs for standalone fuse fiddling
[06:26:08] <LeoNerd> Mmm :)
[06:26:14] <Lambda_Aurigae> use the microsoft way of thinking dude!
[06:27:50] <LeoNerd> "DebugWIRE is enabled wheb DWEN Fuse is programmed and Lock bits are unprogrammed."
[06:28:05] <LeoNerd> I should keep a list of /aaaaaalll/ the typoes I keep finding in AVR datasheets and mail them off
[06:30:20] <Lambda_Aurigae> what's wrong with that?
[06:30:50] <Lambda_Aurigae> DWEN set to 0 and lock bits set to 1 should be correct I would think.
[06:31:17] <Lambda_Aurigae> but, yes, if you find problems, let them know...also, read the erata sheets for the chip you are working with.
[06:38:41] <Guest74251> hello is anybody there?
[06:39:31] <Guest74251> If I connect +9V to AVR I/O closed pin, will it burn or it can stand so high voltage with no current?
[06:39:50] <twnqx> it should be possible if you restrict to 1mA
[06:40:12] <twnqx> but you should be aware that the clamping diodes will short the 9V to the VCC rail
[06:40:12] <Guest74251> did you try it or you just think so?
[06:40:37] <twnqx> no, the chip has clamping giodes, and they are specced for a max current
[06:40:40] <twnqx> diodes*
[06:41:02] <twnqx> so voltages over VCC + diode drop (ca. 0.7V) will be short to VCC
[06:41:04] <apo> clamping geodes
[06:41:26] <Guest74251> thanx! I didn't know about integrated diodes
[06:41:29] <twnqx> i'd probably put a resistor in to make the current as low as the chip will register
[06:41:42] <twnqx> somewhere in the µA
[06:41:59] <Lambda_Aurigae> R T F M...it's in there.
[06:42:26] <Guest74251> I have 9V through 4.4k on I/O pin.. thats about 2mA
[06:42:39] <Guest74251> diodes are zenner i believe?
[06:42:53] <twnqx> no
[06:42:56] <twnqx> regular
[06:43:07] <twnqx> just input to vcc and gnd to input
[06:43:13] <Guest74251> ok so I must calculate 4V through 4.4k yes?
[06:43:25] <twnqx> 9V
[06:43:33] <twnqx> full voltage :)
[06:43:37] <Guest74251> ?
[06:43:47] <Guest74251> it is no logic that diodes go to GND
[06:43:59] <Guest74251> they can go to VCC
[06:44:18] <twnqx> GND --|>|-- IO Pin --|>|-- VCC
[06:44:23] <Guest74251> and if VCC=5 V then voltage across diodes vill be 9-4V
[06:44:30] <Guest74251> 9-5V sorry
[06:44:35] <twnqx> no
[06:44:46] <twnqx> cmos inputs don't absorb current
[06:44:55] <twnqx> so there is no voltage divider
[06:45:07] <Guest74251> not divider
[06:45:22] <Guest74251> GND---> 9V ---> 5V
[06:45:30] <Guest74251> diode go from 9V to 5V
[06:45:41] <Guest74251> not 9V to GND
[06:45:42] <twnqx> the current through the diode will be the same current as through the input pin
[06:45:45] <twnqx> don't play with the voltage
[06:46:39] <Guest74251> ok so I can calculate 4V and 4.4K on diode..
[06:47:05] <twnqx> hmmm
[06:47:15] <twnqx> i would still calculate 9V
[06:47:20] <Guest74251> why?
[06:47:31] <twnqx> because the current is based on the input pin
[06:47:53] <Guest74251> there is 1uA leakage through closed pin
[06:48:04] <Guest74251> but diode gos to VCC
[06:49:15] <Guest74251> that 1mA what you told was through diode?
[06:49:34] <twnqx> i think for all diodes in the chip together
[06:49:39] <twnqx> but i am not sure about that
[06:49:58] <twnqx> http://www.atmel.com/Images/doc2508.pdf is the only place that i quickly found where those didodes are mentioned
[06:50:05] <twnqx> they use 1MOhm in the example
[06:50:51] <twnqx> but that's for direct connection to 240VAC.
[06:51:57] <Guest74251> wtf
[06:52:15] <twnqx> :)
[06:52:50] <Guest74251> 1uA leakage on 5V that is 5/1e-6=5MOhm
[06:53:46] <Guest74251> 310V / 7 MOhm * 5MOhm = 221V on I/O pin???
[06:54:19] <twnqx> liek i said
[06:54:30] <twnqx> the excess voltage is put into the vcc rail
[06:54:44] <twnqx> so better have a regulator that can cope with that :)
[06:55:03] <twnqx> especially if you put in 1mA and your circuit only takes 4-5
[06:55:26] <twnqx> so better get the input current down to 10µA or so
[06:55:59] <Guest74251> that pdf is possible if I/O pin is open
[07:05:52] <twnqx> define "open"
[07:06:01] <twnqx> and "closed"
[07:06:07] <twnqx> in the context of io pins
[07:23:24] <Guest74251> ddrx=1 = open
[07:23:32] <Guest74251> ddrx=0 = closed
[07:23:39] <Guest74251> output mode = open
[07:23:45] <Guest74251> input mode=closed
[07:23:53] <Guest74251> thanx twnqx!!
[07:24:21] <Guest74251> as you mention there will be only 0.7V on i/o pin
[07:24:52] <twnqx> it's for input mode only
[07:24:56] <Guest74251> and current will be less than 1mA (1mA was documented in pdf) so all should work correctly
[07:24:57] <twnqx> output will cause a short
[07:25:15] <twnqx> and most likely destroy the device
[07:25:19] <Guest74251> no
[07:25:40] <twnqx> aynway, you keep the pin as input
[07:26:00] <twnqx> you can believe me on that or take the chance to destroy that device
[07:26:09] <twnqx> that's up to you :)
[07:26:31] <Guest74251> 9V through 4.4K and open i/o will not make U on AVR
[07:26:38] <Guest74251> all voltage will drom on R
[07:26:44] <twnqx> sure
[07:26:57] <twnqx> but you can't read the io pin with it being set to output :P
[07:27:08] <Guest74251> i dont need it
[07:27:12] <twnqx> so ummm
[07:27:20] <twnqx> why d you put 9V to an io pin then? :S
[07:27:23] <Guest74251> i just avare what will happen until ddrx=1
[07:27:36] <twnqx> umm
[07:27:47] <Guest74251> I need to drive some bipolar DAC converter
[07:28:06] <Guest74251> and there is way to produce more than 7V on input
[07:28:17] <twnqx> so you use a 9V pullup
[07:28:21] <twnqx> and drive to 0
[07:28:24] <twnqx> right?
[07:28:28] <Guest74251> yes
[07:28:39] <twnqx> why don't you just drive a normal driver then?
[07:28:49] <Guest74251> i don't have any
[07:28:56] <Guest74251> and shops in my country dont have them
[07:29:15] <Guest74251> one will be available after new year but already have these bipolar devices
[07:29:38] <twnqx> hm
[07:29:42] <Guest74251> so i will not waste time and make my device on bipolar device first
[07:29:56] <Guest74251> ok thanx wery much twnqx :))
[07:29:59] <twnqx> i would probably use a mosfet for that instead of the pullup method
[07:30:02] <Guest74251> you helped me alot
[07:30:10] <Guest74251> 8 pins
[07:30:22] <Guest74251> that will require a lot of space
[07:30:24] <twnqx> but whatever works for you :) just consider that the clock rate might be limited by the currents
[07:30:40] <twnqx> so if it doesn't work at first, try to go a bit slower
[07:30:51] <Guest74251> hmm
[07:31:00] <Guest74251> I hear it first time..
[07:31:15] <Guest74251> what do you mean with clock rate might be limited by the currents
[07:31:19] <Guest74251> ?
[07:31:37] <Guest74251> that is no SPI interface but paralel
[07:31:38] <twnqx> input gate of transistors have a certain charge (parasitic capacitance)
[07:31:54] <Guest74251> o.. ok
[07:32:00] <twnqx> so together with your resistor you end up with an RC filter
[07:32:01] <Guest74251> currents must be fine!
[07:32:16] <twnqx> that's in total a lowpass
[07:32:20] <Guest74251> thanx you a lot one more time. You realy helped me out!! :))
[07:32:23] <jesper--> Hi, newbie here with a very basic question: Why add a crystal to my atmega32? What's the point of adding a crystal to a uC with build in crystal? And if to make it go faster then how fast can you go?
[07:32:40] <twnqx> toggling faster than the RC allows will not make the result register at the DAC
[07:32:42] <twnqx> just telling you :)
[07:33:01] <Guest74251> i understand that twnqx. Thanx!
[07:33:18] <Guest74251> jesper-- AVR have RC not X-tal inside
[07:33:28] <Guest74251> ok bye all!
[07:33:31] <twnqx> cu
[07:33:37] <jesper--> ok, but what's the point?
[07:33:45] <twnqx> speed, and precision
[07:33:48] <Guest74251> x-tal is more precision
[07:33:52] <jesper--> ah ok
[07:33:59] <twnqx> and possibly certain exact clocks you can't reach
[07:34:09] <jesper--> i see
[07:34:17] <twnqx> we were talking about 14.356mhz for precise rs232 clocks recently, for example
[07:34:31] <jesper--> so I guess that there's a "mechanical" limit to how fast a crystal you can add?
[07:34:41] <jesper--> ok
[07:34:51] <twnqx> 16mhz on the datasheet, many devices will run to 20, some to 25
[07:35:05] <twnqx> 8mhz for low voltage varaiants, off the the top of my had
[07:35:11] <twnqx> consult your datasheet :)
[07:35:13] <jesper--> My next project will actually be to add RS232 to my uC. Will I need to use a crystal for that?
[07:35:24] <twnqx> (also, usb megas require an external crytsal for usb operation)
[07:35:28] <jesper--> ok, thanks
[07:35:57] <twnqx> probably not, rs232 is pretty tolerant to variances
[07:36:18] <LeoNerd> If you do software serial properly you can get around that by using noninteger baud divisions
[07:36:23] <jesper--> ok, I just want to send/receive via hyperterminal
[07:36:24] <LeoNerd> The hardware UART doesn't support doing that
[07:36:26] <twnqx> that being said, i almost always use crystals - the internal osciallator mostly is pretty limited
[07:36:41] <twnqx> in clock rate, that is
[07:36:52] <jesper--> ok. I have one. Might as well add it then
[07:37:08] <twnqx> don't forget the capacitors :)
[07:37:47] <jesper--> yeah, my friend drew me a schematic on how to add it. Just need to find it
[07:38:03] <jesper--> thanks guys :)
[07:39:27] <jesper--> sudo make install
[07:39:38] <jesper--> lol oops ^^
[07:39:45] <LeoNerd> Password:
[07:39:49] <jesper--> haha
[07:39:57] <jesper--> ;)
[08:19:41] <Guest74251> twnqx IT's ALIVEEE!!! Thx again!! :D
[08:19:54] <twnqx> :) you're welcome
[16:57:52] <hetii> Hi :)
[17:16:13] <LeoNerd> So... Dilema
[17:16:42] <N1njaneer> Always!
[17:16:43] <LeoNerd> I have a nicely working STK500v2 clone of HVSP working in just over 2K of flash, which should fit nicely on a tiny841 with plenty of space to spare
[17:17:14] <LeoNerd> In that space I'm planning therefore to put some standalone controller logic; give it a few buttons and LEDs to do self-hosted fuse changing - probably just SPI/dW, maybe one or two others.
[17:17:31] <LeoNerd> If I do that though I'll need button and LED control, and therein I run out of IO pins. My dilema is:
[17:18:12] <LeoNerd> I can easily move the LEDs and slower IO control (Vtg/Vprg on/off) to a shift register; e.g. stick a little 74'595 on the board too. That's likely cheapest, but not quite the physically smallest and most compact idea
[17:18:39] <LeoNerd> If instead of a tiny841 I put on something bigger, then I can do the lot with only one chip, but that means going up to something like a 28-pin mega8
[17:19:04] <Tom_itx> 2560 ftw!
[17:19:12] <LeoNerd> That could be said to be overkill ;)
[17:20:15] <LeoNerd> Honestly, it juuust about fits in a tiny841. Of that 14 pin chip, there's power/gnd/reset, and two XTAL pins so 9 IOs left. I need 2 for UART, 4 for high-speed HVSP. 5 remain.
[17:21:41] <LeoNerd> Of those 5 I can put all the buttons in analog read formation on one or maybe two analog in pins, and 3 remaining are the SO/SCK/SS line for the 595
[17:21:49] <LeoNerd> It all fits nicely
[17:23:13] <LeoNerd> Oooor.. maybe I use one as SR SS, and put data/clock reused on the HVSP lines, leaving me with two actually spare pins for future ideas
[17:23:36] <LeoNerd> But anyway: that's my dilema. In summary: small chip + SR for slow output lines, or larger self-contained chip
[17:23:41] <LeoNerd> Of some kind which I haven't worked out yet
[17:24:27] <LeoNerd> Also in entirely unrelated matters: What on earth does anyone use a tiny2313 for? I can't work out what would need -that- many IO pins, and yet fit in that small a flash+RAM footprint
[17:24:35] <Tom_itx> form a comittee and discuss it
[17:24:45] <LeoNerd> I am a comittee. I'm already discussing it :P
[17:24:57] <Tom_itx> i used 2313 for my first programmer
[17:26:50] <Tom_itx> http://tom-itx.no-ip.biz:81/~webpage/attiny2313/mini_usb_batch1.jpg
[17:27:39] <LeoNerd> What's the other chip there?
[17:27:46] <Tom_itx> some buffer chip
[17:28:04] <LeoNerd> Mmm
[17:29:01] <Tom_itx> 74HC125
[17:29:02] <hetii> If I may ask how you need use HVPP instead HVSP ?
[17:29:12] <Tom_itx> depends on the chip
[17:29:19] <LeoNerd> Well, maybe I'll attack my question in a different way: What's a fairly small cheap low-end ATmega chip with a UART, XTAL osc. support and at least 18 IO pins, two of which are analog..?
[17:29:25] <Tom_itx> not all do HVPP or not all do HVSP
[17:29:49] <Tom_itx> LeoNerd, check their parametric table
[17:29:52] <LeoNerd> hetii: Almost all the chips do HVPP; you only find HVSP on the smaller ATtiny chips; the 8 and 14-pin ones which literally don't have enough IO pins to talk HVPP
[17:30:03] <masoudd> analog IO pins?
[17:30:37] <LeoNerd> HVSP appears to me to basically be a dual-lane SPI-like interface which carries the 8 HVPP data lines on one lane, and the control signals on the other lane.
[17:30:57] <LeoNerd> Oddly, for reasons I have absolutely no idea why, it uses an 11-bit clock cycle, with 8 data transfers followed by 3 dead ticks.
[17:31:16] <LeoNerd> (that causes much awkwardness to get Saleae Logic to decode ;) )
[17:31:45] <LeoNerd> masoudd: 2 analog in (ADC); I don't need any DAC.
[17:32:11] <masoudd> Ah, I see
[17:33:05] <hetii> LeoNerd: yep a bit strange.
[17:33:09] <masoudd> I think atmega8 would do
[17:33:31] <Tom_itx> oh god don't use a mega8
[17:33:32] <hetii> I even wonder when its needed. Can imagine when you use reset or set his fusebits wrong. but in other case ?
[17:33:42] <Tom_itx> those things should be put out to die somewhere
[17:33:51] <masoudd> Tom_itx: Why?
[17:33:53] <hetii> I suppose this never happen.
[17:33:55] <LeoNerd> Instead of *dark glasses* put on a die
[17:34:03] <Tom_itx> because the 88 is cheaper and has more features
[17:34:34] <hetii> many of mcu even 32bits are cheaper then atmega8
[17:34:35] <Tom_itx> and the 8 is old
[17:34:41] <LeoNerd> hetii: Having an HVSP burner lets you easily play around with reset disabled giving you one more IO pin - can be useful on the small chips. also I want to play around with debugWire more
[17:34:44] <Tom_itx> as is the 32
[17:34:50] <Tom_itx> and 16
[17:35:14] <masoudd> Tom_itx: why is it still being produced then?
[17:35:16] <hetii> LeoNerd: ture.
[17:35:16] <LeoNerd> Disabling reset on a huge great 64-pin chip doesn't buy you much, relatively. But disabling reset on a tiny85, say, lets you go from 5 GPIOs to 6 and that's a huge difference
[17:35:19] <hetii> *True
[17:35:26] <Tom_itx> ask the atmel gods
[17:35:31] <Tom_itx> only they know
[17:35:42] <LeoNerd> Maybe back-compat?
[17:35:50] <LeoNerd> If people still buy it in large quantities for something
[17:35:53] <Tom_itx> the 88 is backward compatible
[17:35:54] <hetii> but there is another side of atmega8.
[17:35:54] <LeoNerd> Hrm
[17:36:04] <hetii> you have a lot of librarys/project on it
[17:36:26] <hetii> and sometimes you will need waste a lot of time to migrate.
[17:36:47] <Tom_itx> it's time to move on...
[17:36:51] <masoudd> Tom_itx: I'll look into mega88, is there something similar for attiny2313?
[17:36:59] <LeoNerd> ?
[17:37:01] <Tom_itx> no idea
[17:37:16] <Tom_itx> the 48 88 168 328 are all the same series
[17:37:24] <hetii> masoudd: take atmega168 and have fun
[17:37:33] <Tom_itx> agree
[17:38:05] <Tom_itx> http://tom-itx.no-ip.biz:81/~webpage/how_to/atmega168/mega168_howto_main_index.php
[17:38:10] <hetii> even on smd you can easly solder it
[17:38:10] <Tom_itx> start there if you need help with it
[17:39:26] <LeoNerd> Hrm... no SOIC-sized mega88; only a huuuuuge DIP28 or a goddamntinynosureIcouldsolderthis TQFP
[17:39:41] <Tom_itx> weenie
[17:39:45] <Tom_itx> those are easy to solder
[17:39:55] <LeoNerd> Hey.. I only started on 0603s because you told me :P
[17:40:04] <LeoNerd> I think they're about my limit at the moment
[17:40:04] <Tom_itx> and you regret it?
[17:41:10] <hetii> LeoNerd: I used 0402 last time with transfromer solder and don`t cry
[17:41:20] <LeoNerd> Hmm.. Anyway; the TQFPs don't seem much more expensive than a tiny841
[17:41:36] <hetii> http://www.elektroda.pl/rtvforum/viewtopic.php?t=2898488&highlight=
[17:41:58] <LeoNerd> Possibly easier board layout if I consider the lack of having two chips, and associated annoyance of putting the second one on
[17:43:10] <Tom_itx> seriously, they are easy to solder...
[17:46:11] <LeoNerd> Ohwait I remember now: The tiny841 has -two- UARTs, thus giving the possibility of host IO comms on one and debugWire to the target on the other
[17:46:20] <LeoNerd> That's what I was hinking
[17:50:24] <hetii> this tiny841 + ws2812 strip and its possible to do some nice display :)
[17:50:38] <LeoNerd> I think I'll likely stick to tiny841 + 74'595 for my 'v1' attempt, but I'll keep the mega88 in mind as wel
[17:50:52] <LeoNerd> Mm.. I have a single ws2812 at home, need to think of something to do with it
[17:51:02] <LeoNerd> I might make that multi-tap continuity tester I keep pondering
[17:51:16] <hetii> and to control motor:
https://www.youtube.com/watch?v=CWx5ji_sFUw
[17:51:19] <hetii> :)
[18:24:04] <LeoNerd> Ooooh bah. I remember now - the mega88 is literally just a small 8Ki flash version of the 328, which is the chip on the Arduino Nano board I've been deving this on anyway
[18:24:18] <Tom_L> uh huh
[18:24:38] <LeoNerd> Still... TCQP rather than SOIC. They look tiny. :(
[18:24:45] <Tom_L> oh poo
[18:25:01] <LeoNerd> and what the hell is a MLF package?
[18:25:26] <Tom_L> no legs
[18:25:33] <Lambda_Aurigae> LeoNerd, 88/168/328,,,all same chip with different amounts of memory.
[18:25:33] <Tom_L> those are a bit trickier
[18:25:40] <N1njaneer> Basically a QFN
[18:25:51] <N1njaneer> "Micro Leadframe"
[18:26:10] <LeoNerd> Ah.
[18:26:17] * LeoNerd must stop misreading it as MILF, also <.<
[18:26:28] <N1njaneer> Atmel's parts are at least much nicer to work with since they are tapered and the pins generally wrap around the edge a bit.
[18:27:08] <N1njaneer> LeoNerd: Or "MELF" which is the metal electrode leadless face pakages. Most notably those stupid cylindrical diodes.
[18:27:31] <N1njaneer> +packages
[18:28:16] <N1njaneer> QFN can actually be done pretty easily if you are on a budget by very carefully using a heat-gun to reflow the package with solder paste, then doing edge clean-ups on the pads with a generous amount of flux and a mini-wave tip iron.
[18:28:52] <N1njaneer> Obviously an actual reflow oven is a far better way, but the heatgun works extremely well if you are careful to control the temperature and air and not overhead things.
[18:29:09] <LeoNerd> Yah.. I own none of those things :P
[18:29:11] <Tom_L> i replaced U6 & U10 with no heat gun:
http://tom-itx.no-ip.biz:81/~webpage/rue/7I43.jpg
[18:29:26] <Tom_L> mind you it didn't fix the board though
[18:29:50] <Tom_L> oh and U9 but it was easy
[18:30:34] <N1njaneer> Generally you need hot air to rework those parts, especially ones that have a center pad.
[18:30:49] <Tom_L> yeah
[18:31:04] <Tom_L> i got it off by putting it in my toaster
[18:31:04] <LeoNerd> Yeah that's a little further than I think I am currently ;)
[18:31:35] <LeoNerd> I don't spend anywhere enough time or make enough (well, any) money doing electronics stuff to justify much by way of that kind of tooling
[18:31:48] <Tom_L> you will
[18:32:10] <LeoNerd> Maybe... maybe
[18:32:22] <LeoNerd> I could strike lucky and make my millions selling HVSP burners ;)
[18:33:01] <LeoNerd> There's gotta be at least five.. maybe ten people who want one. If I sell them at USD200,000 each, that's easily a million. ;)
[18:33:52] <Tom_L> five that want them that bad?
[18:50:07] <N1njaneer> I haven't ever used HV programming in 12 years of doing stuff with AVR parts :)
[18:50:34] <Lambda_Aurigae> I did once...for the fun of it to see if I could do it with my dragon.
[18:50:39] <Lambda_Aurigae> never needed it though.
[18:51:50] <Lambda_Aurigae> did build a fuse doctor and I use it to reset chips but only ever bricked 2 and was able to recover them with an external clock and ICSP.
[18:53:07] <Lambda_Aurigae> often have kids not follow directions and brick chips then I just use the fuse doctor to reset them.
[18:53:28] <Lambda_Aurigae> but the fuse doctor is an HVPP programmer.
[18:55:15] <Lambda_Aurigae> a real HV programmer would be nice if one was using the reset pin as a gpio though.
[18:55:56] <N1njaneer> What would be really nice is if you could just do something like pull reset to +24V and have the chip automatically just default wipe the fuses back to factory defaults :)
[18:56:44] <LeoNerd> N1njaneer: I'm planning a one-touch button for the HVSP control board for doing exactly that
[18:57:03] * LeoNerd idly wonders if HVSP + two shift registers can achieve HVPP
[18:58:03] <Lambda_Aurigae> probably not.
[18:58:18] <Lambda_Aurigae> I bet the protocols are too different.
[18:58:30] <LeoNerd> They look massively similar to me
[18:58:51] <LeoNerd> I had a quick stare at the SII bit patterns for HVSP vs. the control line bit patterns on HVPP and I thought I could see a correlation of bits
[18:58:58] <LeoNerd> I know the data lines are identical
[18:59:36] <LeoNerd> Also the "OR mask" behaviour to toggle one control line in particular is a total giveaway ;)
[19:00:33] <Lambda_Aurigae> go for it.
[19:40:26] <hetii> time python dump.py: Attached to: /dev/ttyACM0; 2097152/2097152 (100.00%)
[19:41:47] <hetii> times: hardware spi(max speed) 0m12.880s, (bitbang spi) 0m36.542s, (hardware spi 400Khz) 0m48.823s
