I would recommend using a sound generator chip instead of loads of DDS:s. Something like SID, AY-3-8910, or even one of the FM based ones, like the OPN/OPL series of chips. They already support multiple channels, and they're focused on audio output.
Stefan Berndtsson The AY-3-8910 is old and a bit impractical with it's 8-bit parallel bus. Something that uses SPI would be easier to use. I'll check out your suggestions. Cheers.
Julian Ilett True, but the AY-3-8910 bus issue can be overcome with a shift register fairly easily. Since the AY is the one I know well enough (coming from an Atari ST background), it does give you volume and envelope control (individual volume per channel, single envelope shape/freq for the channel set to envelope mode). When outputting audio, you never need to read anything back from it, so it needs 2 pins to control the chip (one of the 3 control pins can be skipped) and 8 for the data. With two chained 74xx595:s you could fully control 4 AY chips, giving you 12 independent channels with only the 3 pins necessary for shifting (all AY:s can share the same 8 bits for data). Or you could put an MCP23S17 in between and get 16 proper IO-pins via SPI.
Those are 5 band resistors first 3 bands are digits, 4th band is multiplier and 5th band is tolerance. Red (2), Red (2), Black (0), Silver (x0.01), Brown (1%); thus 220 * .01 = 2.2 ohm at 1% tolerance. And, the 4k7 is Yellow (4), Purple (7), Black (0), Brown (x10), Brown (1%), thus 470 x 10 = 4.7k ohm at 1%. Tight tolerance resistors (especially those made for military use) are usually coded in 5 band, and a 6th band would be the temperature coefficient.
For a five band resistor the first three bands are the "digits" the third the multiplier, and the last the tolerance. The only difference between a four and five band resistor is the number of significant digits the Ohms are recorded to. The last band is still the tolerance.
For 5 band resistors you have three digits, then the number of zeros and then the tolerance. So the silver band is the number of zeros. In the case of silver its 10^-2 or a 0.01x multiplier so you have 220*10^-2 = 2.2ohm. I think you actually have a gold band which is 10^-1 or 0.1x
In 30 to 50 words...give or take, advise all what this unit does or enables one to do. When that's done, this will become a superb overview, as opposed to now being very good.
The last band on the 5 band resistors is the tolerance, which would be 1%. The AD9850 modules are really good. I've used them before. The module is (or was) cheaper that the IC itself, which is a bit suspicious. The 7th order elliptical filter on board is the LC network you pointed at. There is a problem with these ebay modules though unless the part values have been revised. The AD9850 reference board has a 125MHz clock and a 42MHz filter which is far enough below the 62.5MHz nyquist frequency to avoid aliasing, the AD9851 reference design has a 180MHz clock and a 70MHz filter. These boards have the 125MHz clock but the 70MHz filter from the wrong device's datasheet. This means there is significant aliasing at high frequencies and although the maximum frequency is always quoted as 40MHz based on that being the expected filter frequency, it aliases to buggery in the high frequencies and is unusable above 20MHz. At audio frequencies it shouldn't really make any difference though. Yes the filter is a very high cut off frequency compared to audio, but as long as the output frequency you want is lower than this frequency and the digital is higher than this frequency, it filters it into a near perfect sine whether it's 30mHz or 30MHz The resolution on these is dependant on the clock frequency on both DDS you have. Use a slower clock and each bit will represent a smaller frequency increment, I think you just might have a multiply the frequency by your clock's ratio to 25MHz in software, and if you want an anti-aliasing filter you'd have to redesign that.
KX36 Thanks for the info. The AD9833 looks like a much neater solution and because I'm planning to use it as input to my vocoder, I probably won't need a low pass filter at all. I did see the bit on the data sheet about using a lower frequency crystal to get finer frequency resolution - if 0.1Hz steps is a problem, I'll give that a try. Cheers.
+kx36 +Julian Ilett One thing I've been dealing with is similar to Raduino use, hopefully your familiar with the dds called Raduino by Ashar Farhan of india and other uses, the greek system. A couple of questions. Wondering if the AD dds has less harmonics and would be more easily filtered ? The Si5351A (used in the Raduino, has three different "clock outputs" }supposedly does not have a sine output. The Raduino is shown at hfsignals.com where you can buy a couple of radios which have it , the BITX and uBITX , both great radios. I don't think you can buy the Raduino separately as he is swamped with orders and needs every Raduino he produces. The rotary encoders available (ebay) quite often talk about "20 positions". The one that Ashar uses doesn't have positions, although one rotation may give 20 grey codes. The "20 position" version has mechanical feedback per position. The change of one position is translated into two frequency shifts by the Raduino, which is a pain . So, I'm wondering if either of you have found an encoder similar to the one Ashar uses that doesn't have mechanical feedback ?
Mark Webcraft Alice's eBay listing has all the values listed - lowest is 10Ω. Better get my multimeter... Hmm, 22.5Ω - the silver band remains a mystery. Actually, one of the reasons I bought Alice's pack is because it had a good spread of values. Many of the bulk packs on eBay have a rubbish selection of values - often with half of them below 1k.
Mark Webcraft I think what appeared to be silver is in fact gold, and it's the x0.1 multiplier. It's only on the resistors under 100ohm. Damn those 4-band critters!
Didn't you check your ftdi drivers, or revert to the old ones before using those ftdi clones? or did MS push ut the bugged ones again?, I don't think those are on my machine still, but wonder now?
I am still using the ip65 version of the solar light charge controller to this day with great succes. It is about the same price comparatively. Also, the electrodacus maker recently shared his thoughts on using a 24 volt system with the most common 30 volt mpp solar panels as you can getaway without dc dc conversion and mppt and still get the same efficiency. Maybe that makes for something nice to measure. I think he is on to something here. A 30 amp pwm is about 30 dollars, that is far cheaper then the mppt variant by a factor of 4!
databeestje yes mppt has a larger overhead of cost of the unit, but as with all things will get cheaper, the potential 30 percent (dunno what overall real world is) increase in efficiency is worth having, esp as the tech gets ever cheaper, even with larger panels coming down in price, id still want 30 percent if its cheap to do so
jusb1066 Just to make things clear here, the 30 percent you reffer to is with the common 17 Volt MPP panels on a 12 Volt battery situation. MPPT with DC/DC conversion is a totally valid solution where the PV voltage is very different from the battery voltage. The DC/DC conversion is generally about 95% efficient which makes it worthwhile. That's why I listed the example above, a common 250Wp panel has a ~30 Volt MPP, and about 27-28 when it's warm. That voltage is about the same as the float voltage for a 24 Volt system. The voltage differential is so small that a DC/DC conversion would not give you much more. So, if you want a good performing off grid install on the cheap, use 24 Volts with common 220-250Wp solar panels :)
Hi Julian, I may have missed it somewhere but do you have any more information on that solar controller and also the one it replaced? Are they real PWM controllers? How much no-load current do they take from the battery?
If you really don't want to unbrick them, you can still program all the boards with an ISP programmer. Doesn't really change much in the use of the Arduino IDE (just hold shift while clicking "upload" and it will use the programmer you selected, like an USBasp, instead of the serial connection/bootloader). Or you can still program them via the normal bootloader by connecting an external USB-serial-adapter to the right pins. If you take an adapter that has a DTR pin available (that's the signal line that resets the Arduino to go into the bootloader) it's no different then using the Arduino on-board converter. I'm using USB-serial adapters with the Silabs CP2102 exclusively (have around 10 of them). I looked at the driver history of their chips and the support is at least as good as the FTDI one (minus the hardware bricking part), something you can't say of most other manufacturers (Prolific being a bad example). I've never had any problem with them and in the light of what FTDI did, I'm glad I got them. One of the versions I like best is this one: www.ebay.com/itm/400565980256 $1.60 a piece (or maybe even cheaper in one of the "multipacks"). Btw. you should be able to unbrick the FTDIs without Linux and a command line, at least if you got Windows XP around somewhere. There's a tool on FTDI's website that allows you to reprogram them (never did that, but I read it somewhere). Of course you have to make sure that you don't get the "bricking" driver on you newer Windows again.
superdau When I investigated the FTDI gate affair back in October, I had the safe 2.10 driver. Well, now I have the bricking 2.12 driver and all my fake FTDI chips got bricked in the last 2 days. However, I remember using my FTDI (probably fake) USB/serial converter just recently (RFID522 video 3rd March) and it worked fine. So something appears to have changed quite recently. On FTDI's website, 2.12 appears to be the latest VCP (virtual COM port) driver available. I'll look into this further.
Regarding your penny organ, envelope shaping, intonation, decay and all that, I'm not sure how good this thing is going to sound. Wouldn't a midi project have been better with the arduino and capacitative touch bit acting as the human interface?
Do you have a video on your solar light car battery charger; I didn't see one in you solar playlist. This is exactly what i would like to use to light some low voltage landscape lights and duck coop light in the yard. Would greatly appreciate you making or directing me to an existing more detailed vid.
Julian, I do not want to discourage you but I have bought the same charge controller from Alice and it doesn't do anything except for timer and daylight/night detection. The battery was being charged at 17 V (!!!!) and it didn't enter in float mode, and the radio trick you showed revealed that it didn't absolutely do anything. Now maybe it's just mine that is broken, let me know!! :)
Julian Ilett I guess you can take one of your PWM5s and put it between the solar panel and the solar panel input of Alice's controller so you still have modulation of the voltage (if it is broken like mine) and it should still do timer/daylight detection.
Also, it would be nice if you made a quick video on the math about solar systems. I mean, how to size them according to your load and how to size the batteries, how to calculate how much panels you need, and compare your real setup with the calculated one. There are lots of videos about how to actually make solar panel setups (that's not too hard :D) but none of them ar really great about how to plan them! :)
hmm instead of putting one of those 12$ sound thingies dedicated on each 'note'... Wouldn't it be more cost effective to have only 1 (or 2, 3, ... depending on how many simultaneous notes you want to support) of those thingies and have it produce the tone you want, and 'manage' them from the arduino (so press key, arduino finds a free tone generator, and makes it beep at that tone). Suppose you wanted to make a full 8 octave keyboard, you'd need a whopping amount of those tone generators, but with this approach the amount of keys makes little difference. (well other than having the parts to support the keys ofc).
oreubens Yes, that's exactly how I want to do it. Sorry, I should have made it clearer. I'm thinking of starting with 3 or 4 tone generators and allocating them on a 'last used' basis.
Julian Ilett useless? I love lollipop (once xposed framework is working on ART) But a guy like you should have no problem unlocking the bootloader and flashing kitkat.
Morning Julian, i'm going to buy some 3.7v LiPo bats for a wearable. Can you suggest an Ebay seller to purchase from, looking for quality not quantity. Great channel as always, cheers.
T Komoski Difficult to answer. eBay is awash with fakes, both 18650s and pouch cells. Do you feel comfortable putting lithium cells on airplanes? Buy from a reputable seller in your own country is probably the best advice.
Andy Plater A better move would have been to make the driver run really slowly with the fake chips. That way we'd have got the message, but not completely lost trust in anything FTDI
Did you ever un-brick those Nanos? There's several "how-to" articles. I'm aware of people who un-bricked them and installed Linux to circumvent the problem.
FTDI interesting because after the upset last year, no more was said by anyone. I now have 2.12 driver, supposedly the one that bricks. The file C:\Windows\Inf\setupapi.dev.log shows it was installed on 9th April.2015.
anyone have any figures or best guess for those rubber insulators being better or worse thermally than good old mica insulators. I have to heatsink 15 mosfets and was wondering if i should buy the common rubber type or hunt down the old stye..
There is method for unbricking fake FTDI chips. You can change reference oscillator for ad9833 in order to obtain higher resolution (info in right column on first page of datasheet).
Julian, here's an easy fix for unbricking those Fake FTDI's... ruclips.net/video/RZH_qGautqM/видео.html Just make sure to check for updates on the drivers or they might get bricked again if the older drivers are still present.
jusb1066 Just tried that, but it's not working. Mprog cannot see my bricked Nanos. I have one Nano that appears to have a genuine FT232 - Mprog can see that one.
Julian Ilett You might also have a wrong Vendor ID on those chips... Here are some clues: ok1zia.nagano.cz/wiki/Programming_FTDI_devices Basically, you might have to edit the driver's files manually so that Mprog recognizes other Vendor ID/Product ID combinations. Then Mprog should detect the fakes and allow you to change the VID/PID.
Julian Ilett Ok, just found the guide I was looking for. I've followed these steps when mine got bricked: forum.gsmhosting.com/vbb/f799/ft232r-repair-official-guide-1875023/
You may be able to unbrick your FT232s: ruclips.net/video/RZH_qGautqM/видео.html Although you'd probably have to do something fancy like install a linux virtual machine and use arduino through it to stop them getting bricked again, just a thought!
FTDI has pulled their insidious driver from windows update, so this should not happen again - if you replace the bad driver with the current one. The bricked ones can be fixed with linux: www.minipwner.com/index.php/unbrickftdi000
***** I downloaded 2.12 from FTDI & installed it on XP & it didn't brick a very cheap "FTDI" module. Can't say for sure if it's a clone though.
9 лет назад
***** Triple bricked and unbricked here. First time you connect it works, next time it already have the 0x0000 PID. Windows 8.1 clean instalation this month. Pretty sure the thing is back again!
Great video. I believe it has the weird recessed 7 segments so that in very sunny conditions you can still read the display.
Dan Nesnidal Could be, I was thinking maybe originally it was meant to be socketed and some costcutter got to it...
+Dan Nesnidal I thought the same when I saw that too.
I would recommend using a sound generator chip instead of loads of DDS:s. Something like SID, AY-3-8910, or even one of the FM based ones, like the OPN/OPL series of chips. They already support multiple channels, and they're focused on audio output.
Stefan Berndtsson The AY-3-8910 is old and a bit impractical with it's 8-bit parallel bus. Something that uses SPI would be easier to use. I'll check out your suggestions. Cheers.
Julian Ilett True, but the AY-3-8910 bus issue can be overcome with a shift register fairly easily. Since the AY is the one I know well enough (coming from an Atari ST background), it does give you volume and envelope control (individual volume per channel, single envelope shape/freq for the channel set to envelope mode). When outputting audio, you never need to read anything back from it, so it needs 2 pins to control the chip (one of the 3 control pins can be skipped) and 8 for the data. With two chained 74xx595:s you could fully control 4 AY chips, giving you 12 independent channels with only the 3 pins necessary for shifting (all AY:s can share the same 8 bits for data). Or you could put an MCP23S17 in between and get 16 proper IO-pins via SPI.
Awesome video Julian! Bit slow on watching it, but finally :) Cant wait for your summer projects! Keep up the great work!
Those are 5 band resistors first 3 bands are digits, 4th band is multiplier and 5th band is tolerance. Red (2), Red (2), Black (0), Silver (x0.01), Brown (1%); thus 220 * .01 = 2.2 ohm at 1% tolerance. And, the 4k7 is Yellow (4), Purple (7), Black (0), Brown (x10), Brown (1%), thus 470 x 10 = 4.7k ohm at 1%.
Tight tolerance resistors (especially those made for military use) are usually coded in 5 band, and a 6th band would be the temperature coefficient.
For a five band resistor the first three bands are the "digits" the third the multiplier, and the last the tolerance. The only difference between a four and five band resistor is the number of significant digits the Ohms are recorded to. The last band is still the tolerance.
For 5 band resistors you have three digits, then the number of zeros and then the tolerance. So the silver band is the number of zeros. In the case of silver its 10^-2 or a 0.01x multiplier so you have 220*10^-2 = 2.2ohm. I think you actually have a gold band which is 10^-1 or 0.1x
Just a thought, why not use 555 timers for each note of your organ? Much cheaper and more than enough for the frequency range you're going to need.
In 30 to 50 words...give or take, advise all what this unit does or enables one to do. When that's done, this will become a superb overview, as opposed to now being very good.
The last band on the 5 band resistors is the tolerance, which would be 1%.
The AD9850 modules are really good. I've used them before. The module is (or was) cheaper that the IC itself, which is a bit suspicious. The 7th order elliptical filter on board is the LC network you pointed at. There is a problem with these ebay modules though unless the part values have been revised. The AD9850 reference board has a 125MHz clock and a 42MHz filter which is far enough below the 62.5MHz nyquist frequency to avoid aliasing, the AD9851 reference design has a 180MHz clock and a 70MHz filter. These boards have the 125MHz clock but the 70MHz filter from the wrong device's datasheet. This means there is significant aliasing at high frequencies and although the maximum frequency is always quoted as 40MHz based on that being the expected filter frequency, it aliases to buggery in the high frequencies and is unusable above 20MHz.
At audio frequencies it shouldn't really make any difference though. Yes the filter is a very high cut off frequency compared to audio, but as long as the output frequency you want is lower than this frequency and the digital is higher than this frequency, it filters it into a near perfect sine whether it's 30mHz or 30MHz
The resolution on these is dependant on the clock frequency on both DDS you have. Use a slower clock and each bit will represent a smaller frequency increment, I think you just might have a multiply the frequency by your clock's ratio to 25MHz in software, and if you want an anti-aliasing filter you'd have to redesign that.
KX36 Thanks for the info. The AD9833 looks like a much neater solution and because I'm planning to use it as input to my vocoder, I probably won't need a low pass filter at all. I did see the bit on the data sheet about using a lower frequency crystal to get finer frequency resolution - if 0.1Hz steps is a problem, I'll give that a try. Cheers.
+kx36 +Julian Ilett
One thing I've been dealing with is similar to Raduino use, hopefully your familiar with the dds called Raduino by Ashar Farhan of india and other uses, the greek system. A couple of questions.
Wondering if the AD dds has less harmonics and would be more easily filtered ? The Si5351A (used in the Raduino, has three different "clock outputs" }supposedly does not have a sine output. The Raduino is shown at hfsignals.com where you can buy a couple of radios which have it , the BITX and uBITX , both great radios. I don't think you can buy the Raduino separately as he is swamped with orders and needs every Raduino he produces.
The rotary encoders available (ebay) quite often talk about "20 positions". The one that Ashar uses doesn't have positions, although one rotation may give 20 grey codes. The "20 position" version has mechanical feedback per position. The change of one position is translated into two frequency shifts by the Raduino, which is a pain .
So, I'm wondering if either of you have found an encoder similar to the one Ashar uses that doesn't have mechanical feedback ?
Resistors, value value value multiplier tolerance so red red black silver brown is 22 ohm 1%. I'm assuming you've worked it out by now though
@TMM The resistor "Red Red Black Silver Brown" is a 2.2 ohm not 22 ohm
220 x .01 @ 1%.
Gold band is x .1
Mark Webcraft Alice's eBay listing has all the values listed - lowest is 10Ω. Better get my multimeter... Hmm, 22.5Ω - the silver band remains a mystery. Actually, one of the reasons I bought Alice's pack is because it had a good spread of values. Many of the bulk packs on eBay have a rubbish selection of values - often with half of them below 1k.
Mark Webcraft I think what appeared to be silver is in fact gold, and it's the x0.1 multiplier. It's only on the resistors under 100ohm. Damn those 4-band critters!
Didn't you check your ftdi drivers, or revert to the old ones before using those ftdi clones? or did MS push ut the bugged ones again?, I don't think those are on my machine still, but wonder now?
I am still using the ip65 version of the solar light charge controller to this day with great succes. It is about the same price comparatively.
Also, the electrodacus maker recently shared his thoughts on using a 24 volt system with the most common 30 volt mpp solar panels as you can getaway without dc dc conversion and mppt and still get the same efficiency. Maybe that makes for something nice to measure. I think he is on to something here. A 30 amp pwm is about 30 dollars, that is far cheaper then the mppt variant by a factor of 4!
databeestje yes mppt has a larger overhead of cost of the unit, but as with all things will get cheaper, the potential 30 percent (dunno what overall real world is) increase in efficiency is worth having, esp as the tech gets ever cheaper, even with larger panels coming down in price, id still want 30 percent if its cheap to do so
jusb1066 Just to make things clear here, the 30 percent you reffer to is with the common 17 Volt MPP panels on a 12 Volt battery situation. MPPT with DC/DC conversion is a totally valid solution where the PV voltage is very different from the battery voltage. The DC/DC conversion is generally about 95% efficient which makes it worthwhile.
That's why I listed the example above, a common 250Wp panel has a ~30 Volt MPP, and about 27-28 when it's warm. That voltage is about the same as the float voltage for a 24 Volt system. The voltage differential is so small that a DC/DC conversion would not give you much more.
So, if you want a good performing off grid install on the cheap, use 24 Volts with common 220-250Wp solar panels :)
databeestje good reply and info ^
Hi Julian, I may have missed it somewhere but do you have any more information on that solar controller and also the one it replaced? Are they real PWM controllers? How much no-load current do they take from the battery?
Hmm, I wonder if that AD9850 can be used as a VFO for ham radio applications. Could be stable ( and accurate ) enough.
If you really don't want to unbrick them, you can still program all the boards with an ISP programmer. Doesn't really change much in the use of the Arduino IDE (just hold shift while clicking "upload" and it will use the programmer you selected, like an USBasp, instead of the serial connection/bootloader).
Or you can still program them via the normal bootloader by connecting an external USB-serial-adapter to the right pins. If you take an adapter that has a DTR pin available (that's the signal line that resets the Arduino to go into the bootloader) it's no different then using the Arduino on-board converter.
I'm using USB-serial adapters with the Silabs CP2102 exclusively (have around 10 of them). I looked at the driver history of their chips and the support is at least as good as the FTDI one (minus the hardware bricking part), something you can't say of most other manufacturers (Prolific being a bad example). I've never had any problem with them and in the light of what FTDI did, I'm glad I got them.
One of the versions I like best is this one:
www.ebay.com/itm/400565980256
$1.60 a piece (or maybe even cheaper in one of the "multipacks").
Btw. you should be able to unbrick the FTDIs without Linux and a command line, at least if you got Windows XP around somewhere. There's a tool on FTDI's website that allows you to reprogram them (never did that, but I read it somewhere). Of course you have to make sure that you don't get the "bricking" driver on you newer Windows again.
superdau When I investigated the FTDI gate affair back in October, I had the safe 2.10 driver. Well, now I have the bricking 2.12 driver and all my fake FTDI chips got bricked in the last 2 days. However, I remember using my FTDI (probably fake) USB/serial converter just recently (RFID522 video 3rd March) and it worked fine. So something appears to have changed quite recently. On FTDI's website, 2.12 appears to be the latest VCP (virtual COM port) driver available. I'll look into this further.
Regarding your penny organ, envelope shaping, intonation, decay and all that, I'm not sure how good this thing is going to sound. Wouldn't a midi project have been better with the arduino and capacitative touch bit acting as the human interface?
Nice video but what happened to the "EBIKE" ??
Andrew Bryant pianotunerbristol Tomasz Schafernaker has put a stop to that for the time being
Why?
Do you have a video on your solar light car battery charger; I didn't see one in you solar playlist. This is exactly what i would like to use to light some low voltage landscape lights and duck coop light in the yard. Would greatly appreciate you making or directing me to an existing more detailed vid.
This is like very HQ generator used for driving heat induction or RF stuff. Too fancy for just audio project.
Julian, I do not want to discourage you but I have bought the same charge controller from Alice and it doesn't do anything except for timer and daylight/night detection. The battery was being charged at 17 V (!!!!) and it didn't enter in float mode, and the radio trick you showed revealed that it didn't absolutely do anything.
Now maybe it's just mine that is broken, let me know!!
:)
MrOpenGL That's all I want it to do really, but 17v is not too clever. I'll check all these things when I connect it up.
Julian Ilett I guess you can take one of your PWM5s and put it between the solar panel and the solar panel input of Alice's controller so you still have modulation of the voltage (if it is broken like mine) and it should still do timer/daylight detection.
Also, it would be nice if you made a quick video on the math about solar systems. I mean, how to size them according to your load and how to size the batteries, how to calculate how much panels you need, and compare your real setup with the calculated one. There are lots of videos about how to actually make solar panel setups (that's not too hard :D) but none of them ar really great about how to plan them! :)
Is it possible to buy the thermal heat transfer pads ??
hmm instead of putting one of those 12$ sound thingies dedicated on each 'note'... Wouldn't it be more cost effective to have only 1 (or 2, 3, ... depending on how many simultaneous notes you want to support) of those thingies and have it produce the tone you want, and 'manage' them from the arduino (so press key, arduino finds a free tone generator, and makes it beep at that tone).
Suppose you wanted to make a full 8 octave keyboard, you'd need a whopping amount of those tone generators, but with this approach the amount of keys makes little difference. (well other than having the parts to support the keys ofc).
oreubens Yes, that's exactly how I want to do it. Sorry, I should have made it clearer. I'm thinking of starting with 3 or 4 tone generators and allocating them on a 'last used' basis.
Hey a fantastic app for electronics hobbyists is electrodroid. It has tons of tools, calculators, ect. You should try it on your nexus 7
***** Unfortunately, my Nexus 7 is near useless now it has Lollipop.
Julian Ilett useless? I love lollipop (once xposed framework is working on ART)
But a guy like you should have no problem unlocking the bootloader and flashing kitkat.
Morning Julian, i'm going to buy some 3.7v LiPo bats for a wearable. Can you suggest an Ebay seller to purchase from, looking for quality not quantity. Great channel as always, cheers.
T Komoski Difficult to answer. eBay is awash with fakes, both 18650s and pouch cells. Do you feel comfortable putting lithium cells on airplanes? Buy from a reputable seller in your own country is probably the best advice.
Steven Bauer Thanks, at least they ship from USA.
Update your windows drivers for the FTDI chip - i'm sure they updated them in response to the outrage at them bricking our stuff...
Sure the FTDI's are bricked? Just use a old driver and they should work. (The new Driver detect's fake chips and deny the work but don't brick them)
It's sad that FTDI has basically destroyed their customer-base who were only barely hanging on to begin with due to the price of the chips.
Andy Plater A better move would have been to make the driver run really slowly with the fake chips. That way we'd have got the message, but not completely lost trust in anything FTDI
Julian Ilett or pop up an intrusive message
Did you ever un-brick those Nanos? There's several "how-to" articles. I'm aware of people who un-bricked them and installed Linux to circumvent the problem.
do not throw out the bricked fdi chips you can unbrick them i have
regards john
FTDI bastages. You aren't the only one switching brands. You can un-brick those with a bit of fiddling.
@2:26 MOSFETS are also Transistors, right?
metal-oxide-semiconductor field-effect transistor
There is a synth project for the RASPI....
Gary Hunkin Finally, a use for my Raspberry Pi!
FTDI interesting because after the upset last year, no more was said by anyone. I now have 2.12 driver, supposedly the one that bricks. The file C:\Windows\Inf\setupapi.dev.log shows it was installed on 9th April.2015.
David Pilling That date fits with my observations. I wonder if that was part of a recent Windows update.
anyone have any figures or best guess for those rubber insulators being better or worse thermally than good old mica insulators.
I have to heatsink 15 mosfets and was wondering if i should buy the common rubber type or hunt down the old stye..
There is method for unbricking fake FTDI chips. You can change reference oscillator for ad9833 in order to obtain higher resolution (info in right column on first page of datasheet).
rapsod1911 I'll see how the 0.1Hz resolution works out first. Cheers.
will I get volt in negative. is output ac ?
Julian, here's an easy fix for unbricking those Fake FTDI's...
ruclips.net/video/RZH_qGautqM/видео.html
Just make sure to check for updates on the drivers or they might get bricked again if the older drivers are still present.
Tiago Costa good find, and from windows too, Julian, worth a go?
jusb1066 Just tried that, but it's not working. Mprog cannot see my bricked Nanos. I have one Nano that appears to have a genuine FT232 - Mprog can see that one.
Julian Ilett You might also have a wrong Vendor ID on those chips... Here are some clues: ok1zia.nagano.cz/wiki/Programming_FTDI_devices
Basically, you might have to edit the driver's files manually so that Mprog recognizes other Vendor ID/Product ID combinations. Then Mprog should detect the fakes and allow you to change the VID/PID.
Julian Ilett Ok, just found the guide I was looking for. I've followed these steps when mine got bricked: forum.gsmhosting.com/vbb/f799/ft232r-repair-official-guide-1875023/
Tiago Costa This guide worked for me: ljck.org/unbrick_ftdi
Does android brick them? or FTDI company?
Do ad9850 only generates the wave on positive part of the voltage axis? Or does it can produce the wave either side of the axis ?
positive side only
Julian Ilett thank you. Is there any Ic that can generate sinewave either side of the voltage axis? Plz let me know.
can I buy a broken Arduino from you?
I call the language in that "manual" "Chinglish".
You may be able to unbrick your FT232s:
ruclips.net/video/RZH_qGautqM/видео.html
Although you'd probably have to do something fancy like install a linux virtual machine and use arduino through it to stop them getting bricked again, just a thought!
FTDI has pulled their insidious driver from windows update, so this should not happen again - if you replace the bad driver with the current one.
The bricked ones can be fixed with linux: www.minipwner.com/index.php/unbrickftdi000
***** I managed to unbrick the fake FTDIs using this guide: ljck.org/unbrick_ftdi
***** The driver (2.12) is back on windows update. Clones are getting bricked again.
***** I downloaded 2.12 from FTDI & installed it on XP & it didn't brick a very cheap "FTDI" module. Can't say for sure if it's a clone though.
***** Triple bricked and unbricked here. First time you connect it works, next time it already have the 0x0000 PID. Windows 8.1 clean instalation this month. Pretty sure the thing is back again!
to buy cheaper on ebay change the region langage of mozilla to US and go to ebay.com
あなたは日本のお金（50円玉）をなんてことに使ってるんだっ!! 許されない行為!!!!