On Tue, Dec 9, 2014 at 8:49 AM, joem joem@martindale-electric.co.uk wrote:
On Mon, 2014-12-08 at 15:45 +0000, Luke Kenneth Casson Leighton wrote:
... *deep breath*... but.... hold everything, i've never been a great fan of the 1024x600 LCD panels
hmmm... thinks odroid-vu http://www.hardkernel.com/main/products/prdt_info.php?g_code=G140383714860
Miniature HDMI monitor for $120. I bought one. If there were more of these things around with variety then lessens need for fixed screen.
which reminds me: i need to work out a circuit for driving the LCD backlight. does anyone have e.g. one of these: http://www.adafruit.com/products/1931
i need an example circuit to work from: the one that i have uses the SY7201 and it is designed to pulse the LEDs, whereas the Hannstar LCDs have their own dedicated LED backlight PWM input.
l.
On Tue, 2014-12-09 at 14:33 +0000, Luke Kenneth Casson Leighton wrote:
On Tue, Dec 9, 2014 at 8:49 AM, joem joem@martindale-electric.co.uk wrote:
On Mon, 2014-12-08 at 15:45 +0000, Luke Kenneth Casson Leighton wrote:
... *deep breath*... but.... hold everything, i've never been a great fan of the 1024x600 LCD panels
hmmm... thinks odroid-vu http://www.hardkernel.com/main/products/prdt_info.php?g_code=G140383714860
Miniature HDMI monitor for $120. I bought one. If there were more of these things around with variety then lessens need for fixed screen.
which reminds me: i need to work out a circuit for driving the LCD backlight. does anyone have e.g. one of these: http://www.adafruit.com/products/1931
I have similar but 5" from Aliexpress (used in reversing cars). When I try to replicate the board, came across this idea that there is some 'firmware' like parameters that have to be set up to make the chip and LCD work correctly. There was no further information about how one might go about doing that.
i need an example circuit to work from: the one that i have uses the SY7201 and it is designed to pulse the LEDs, whereas the Hannstar LCDs have their own dedicated LED backlight PWM input.
Not sure what that means. The example circuit is given in their PDF. (The inductor is worked out by swapping out some reasonable values until its just right which you can get by looking at the scope signals, or do some calculations to estimate best value.) The inductance changes with changes to LED chain. The LED chain is electrically isolated from the rest of the display - so nothing to worry about except burnt out LEDs. Net current down LED chain normally not exceed 20mA. That is how the limiting resistor is set.
(I normally use a 30V supply and limit the current to 20mA by having a serial resistor and upping the voltage until good brightness is reached. Obviously not clever if it has to work from batteries - but it is a starting point to work out the net LED voltage that has to be subtracted from calculations to get limiting resistor, currents, and inductance.)
On Tue, Dec 9, 2014 at 4:05 PM, joem joem@martindale-electric.co.uk wrote:
On Tue, 2014-12-09 at 14:33 +0000, Luke Kenneth Casson Leighton wrote:
On Tue, Dec 9, 2014 at 8:49 AM, joem joem@martindale-electric.co.uk wrote:
On Mon, 2014-12-08 at 15:45 +0000, Luke Kenneth Casson Leighton wrote:
... *deep breath*... but.... hold everything, i've never been a great fan of the 1024x600 LCD panels
hmmm... thinks odroid-vu http://www.hardkernel.com/main/products/prdt_info.php?g_code=G140383714860
Miniature HDMI monitor for $120. I bought one. If there were more of these things around with variety then lessens need for fixed screen.
which reminds me: i need to work out a circuit for driving the LCD backlight. does anyone have e.g. one of these: http://www.adafruit.com/products/1931
I have similar but 5" from Aliexpress (used in reversing cars). When I try to replicate the board, came across this idea that there is some 'firmware' like parameters that have to be set up to make the chip and LCD work correctly. There was no further information about how one might go about doing that.
there's a guy who has managed exactly that [reverse-engineering]... let me find his paaage.... ah here we go: http://tech.mattmillman.com/lcd/rm5251/
he created a tool that modifies the binary so that you can select whatever you want.
if you have one of those boards can i ask you the favour of locating the white LED circuit and sending me a macro-zoom photo (both sides) of that section of the board, highest resolution you can?
i need an example circuit to work from: the one that i have uses the SY7201 and it is designed to pulse the LEDs, whereas the Hannstar LCDs have their own dedicated LED backlight PWM input.
Not sure what that means.
pin 30 of the HSD101PWW datasheet is "LED PWM brightness".
The example circuit is given in their PDF.
i found the application note: http://www.elecinfo.com/upload/community/2013/12/30/1388365231-20314.pdf
the circuit is *radically* different from the [proven, working] one that has been in circulation for years. it involves an RC circuit with a link into the FB: EN/PWM is used purely for "EN" and the FB line is given an extra "kick" periodically via a PWM to fool it into switching off.
all that has to go, which leaves me in a position to need to understand how the circuit (that i copied) works, and adapt it.
... or find an alternative.
(I normally use a 30V supply and limit the current to 20mA by having a serial resistor and upping the voltage until good brightness is reached.
unfortunately i don't have the luxury of being able to experiment like that. remember that i am endeavouring to *establish* this business, i am not operating from the position of having an R&D budget from established business funds and established sales in order to make these kinds of experiments.
so my preferred path is to use a pre-existing proven circuit. this is proving hard to track down.
last resort i will make something [conservative] and test it. my primary goal here is to get a working prototype that can begin a 2nd crowdfunding campaign.
Obviously not clever if it has to work from batteries
yes, this does.
- but it
is a starting point to work out the net LED voltage that has to be subtracted from calculations to get limiting resistor, currents, and inductance.)
ack. thx joe.
On Tue, 2014-12-09 at 17:21 +0000, Luke Kenneth Casson Leighton wrote:
... *deep breath*... but.... hold everything, i've never been a great fan of the 1024x600 LCD panels
hmmm... thinks odroid-vu http://www.hardkernel.com/main/products/prdt_info.php?g_code=G140383714860
Miniature HDMI monitor for $120. I bought one. If there were more of these things around with variety then lessens need for fixed screen.
which reminds me: i need to work out a circuit for driving the LCD backlight. does anyone have e.g. one of these: http://www.adafruit.com/products/1931
I have similar but 5" from Aliexpress (used in reversing cars). When I try to replicate the board, came across this idea that there is some 'firmware' like parameters that have to be set up to make the chip and LCD work correctly. There was no further information about how one might go about doing that.
there's a guy who has managed exactly that [reverse-engineering]... let me find his paaage.... ah here we go: http://tech.mattmillman.com/lcd/rm5251/
he created a tool that modifies the binary so that you can select whatever you want.
if you have one of those boards can i ask you the favour of locating the white LED circuit and sending me a macro-zoom photo (both sides) of that section of the board, highest resolution you can?
Will try
i need an example circuit to work from: the one that i have uses the SY7201 and it is designed to pulse the LEDs, whereas the Hannstar LCDs have their own dedicated LED backlight PWM input.
Not sure what that means.
pin 30 of the HSD101PWW datasheet is "LED PWM brightness".
The example circuit is given in their PDF.
i found the application note: http://www.elecinfo.com/upload/community/2013/12/30/1388365231-20314.pdf
the circuit is *radically* different from the [proven, working] one that has been in circulation for years. it involves an RC circuit with a link into the FB: EN/PWM is used purely for "EN" and the FB line is given an extra "kick" periodically via a PWM to fool it into switching off.
all that has to go, which leaves me in a position to need to understand how the circuit (that i copied) works, and adapt it.
From datasheet:
"SY7201:Enable and dimming control. When used as enable input, pull high to turn on IC. When used as dimming input, the first pulse should be longer than 200ns to turn on IC. And the recommend dimming frequency range is 20kHz~1MHz.
SY7201A: Enable and dimming control. When used as enable input, pull high to turn on IC. When used as dimming input, the first pulse should be longer than 150us to turn on IC. And the recommend dimming frequency range is 100Hz~10kHz."
So its saying if EN is set high, it will turn on the LED to full brightness. : If instead you were to send a PWM signal say 1kHz for SY7201A then LED can be dimmed depending on MARK:SPACE ratio of the signal. : RC circuit not needed. : RC circuit can be set up to pull EN line high to turn LCD on to max brightness if the PWM source (the CPU) failed to boot.
On Wed, Dec 10, 2014 at 7:53 AM, joem joem@martindale-electric.co.uk wrote:
if you have one of those boards can i ask you the favour of locating the white LED circuit and sending me a macro-zoom photo (both sides) of that section of the board, highest resolution you can?
Will try
star.
If instead you were to send a PWM signal say 1kHz for SY7201A then LED can be dimmed depending on MARK:SPACE ratio of the signal. : RC circuit not needed.
the only reason i can think of that it's been done is that the RC circuit can be used, at high enough frequency, to vary the current rather than do actual PWM at coarse/low frequency.
anyway, it's gone.
l.
ok well it looks like that 7in hannstar LCD simply takes a 3.3v supply for the backlight :) the 10in one needs the SY7201, which is good news all round.
l.
arm-netbook@lists.phcomp.co.uk
-
joem -
Luke Kenneth Casson Leighton