OctoWS2811 with WS2815 leds - 12v DC
- Thread starter bpenchev
- Start date
- Status
PaulStoffregen
Well-known member
But do they *really* take a 12V data signal?
So far, every 12V addressable LED product we've seen actually takes a 5V logic signal, not 12V. They use 12V power, but NOT 12V data. These are unlikely to be any different. But if they are, please give us a link to the product page, datasheet or other info about the signal they require.
Or post a link and someone here might be able to help. Or, other forum readers might learn something.
PaulStoffregen
Well-known member
Maybe try asking the vendor? It is their product, afterall....
Yeah, I know that's probably fruitless. Even if they do reply, how likely is their answer to be right when the datasheet is filled with obviously contradictory info?
But it's always interesting to hear what they say, if they even bother to reply. Please let us know?
Yeah, I know that's probably fruitless. Even if they do reply, how likely is their answer to be right when the datasheet is filled with obviously contradictory info?
But it's always interesting to hear what they say, if they even bother to reply. Please let us know?
Thanks for the link.
The question is, "What voltage levels are required on the digital inputs for a logical Low and High?".
See the section: Electrical Characteristics
High-level Input: 0.7 x VDD
Low-level Input: 0.3 x VDD
For both, the required voltages are relative to VDD. Therefore, if VDD is 12 V, then >= 8.4 V is required for a logical high and <= 3.6 V is required for a logical low.
@VDD == 12 V
High-level Input: >= 8.4 V
Low-level Input: <= 3.6 V
This is the spec you must meet for the LEDs.
A next step would be to determine the High-level and Low-Level Output voltages of the level shifters on the OctoWS2811 (74HCT245).
It would then be known exactly what voltages must be translated from and to.
[Edit]
SN74HCT245, IOH, IOL = –20 μA
High-level Output: >= 4.4 V
Low-Level Output: <= 0.1 V
The question is, "What voltage levels are required on the digital inputs for a logical Low and High?".
See the section: Electrical Characteristics
High-level Input: 0.7 x VDD
Low-level Input: 0.3 x VDD
For both, the required voltages are relative to VDD. Therefore, if VDD is 12 V, then >= 8.4 V is required for a logical high and <= 3.6 V is required for a logical low.
@VDD == 12 V
High-level Input: >= 8.4 V
Low-level Input: <= 3.6 V
This is the spec you must meet for the LEDs.
A next step would be to determine the High-level and Low-Level Output voltages of the level shifters on the OctoWS2811 (74HCT245).
It would then be known exactly what voltages must be translated from and to.
[Edit]
SN74HCT245, IOH, IOL = –20 μA
High-level Output: >= 4.4 V
Low-Level Output: <= 0.1 V
Last edited:
PaulStoffregen
Well-known member
Is that the reply from the vendor? Or is it your reading of the contradictory info in the datasheet?
This is the part used on the Octo boards PJRC sells. Fairchild's datasheet is very reliable.
http://www.onsemi.com/PowerSolutions/product.do?id=MM74HCT245
They have numbers for minimal load, 6mA and 7.2mA, and 3 temperature ranges. For the light load case, the output is within 0.1 volt of GND and VCC, which is connected to VIN-VUSB on Teensy, so nominally 5V but does of course depend on whatever you're using to power Teensy.
This is the part used on the Octo boards PJRC sells. Fairchild's datasheet is very reliable.
http://www.onsemi.com/PowerSolutions/product.do?id=MM74HCT245
They have numbers for minimal load, 6mA and 7.2mA, and 3 temperature ranges. For the light load case, the output is within 0.1 volt of GND and VCC, which is connected to VIN-VUSB on Teensy, so nominally 5V but does of course depend on whatever you're using to power Teensy.
Here is another China manufacturer products.
WS2815A is listed as Input Voltage 5V, and when you download the specs it is set as 12V
Thanks for all response. I will try to order them in US, if i can find them.
Otherwise will have to wait for them to come from China. I will let you know when i have my hands on them.
WS2815A is listed as Input Voltage 5V, and when you download the specs it is set as 12V
Thanks for all response. I will try to order them in US, if i can find them
.Otherwise will have to wait for them to come from China. I will let you know when i have my hands on them.
WS2815 and GS8208 both are 12v addressable led.
WS2815 led datasheet:http://www.normandled.com/upload/201808/WS2815 LED Datasheet.pdf
GS8208 led datasheet:http://www.normandled.com/upload/201805/GS8208 LED Datasheet.pdf
WS2815 led datasheet:http://www.normandled.com/upload/201808/WS2815 LED Datasheet.pdf
GS8208 led datasheet:http://www.normandled.com/upload/201805/GS8208 LED Datasheet.pdf
I found this thread a while ago because I was wondering the same thing and in the meantime I have already tried it myself and I can confirm:
The datasheets are confusing. Even though they use Vdd for both logic and power, VDD=4.5～5.5V seems to be for logic and VDD=+9.5~+13.5 for power in case of the WS2815. I was able to use both WS2815 and GS8208 without problems using a 5V logic signal. No idea about the internals though.
Using the same trick as with 5V LED drivers to supply the first LED with a lower voltage with the help of a voltage divider to use it as a logic lvl shifter has not worked as well/consistently as in case of e.g. WS2812, thus I used a logic-level shifter from 3.3V to 5V to drive it from a 3.3V chip when necessary.
To sum up, you seem to be able to supply them with 12V and drive them with 5V level logic (e.g. arduino or esp2866 & 3.3V to 5V logic level shifting as with the SN74LVC1T45DBVR).
The datasheets are confusing. Even though they use Vdd for both logic and power, VDD=4.5～5.5V seems to be for logic and VDD=+9.5~+13.5 for power in case of the WS2815. I was able to use both WS2815 and GS8208 without problems using a 5V logic signal. No idea about the internals though.
Using the same trick as with 5V LED drivers to supply the first LED with a lower voltage with the help of a voltage divider to use it as a logic lvl shifter has not worked as well/consistently as in case of e.g. WS2812, thus I used a logic-level shifter from 3.3V to 5V to drive it from a 3.3V chip when necessary.
To sum up, you seem to be able to supply them with 12V and drive them with 5V level logic (e.g. arduino or esp2866 & 3.3V to 5V logic level shifting as with the SN74LVC1T45DBVR).
PaulStoffregen
Well-known member
I managed to contact someone from World Semi (on Facebook, of all places...) and they confirmed "VDD=4.5～5.5V" is a mistake in the WS2815 datasheet. They said it will be fixed. No idea of the time frame.
I'm curious to see if whoever edits the datasheet merely replaces the VDD info with the spec from another page, or if they thoughfully update the voltage levels in that table which are currently expressed as ratios of VDD.
I'm curious to see if whoever edits the datasheet merely replaces the VDD info with the spec from another page, or if they thoughfully update the voltage levels in that table which are currently expressed as ratios of VDD.
Looks like the IC runs at 5V
I have a roll of ws2815 LED on my desk.
Win #1:
They do in fact run with 5V logic input. I actually just wired the GND/DAT from the end of a normal neopixel (ws2812) strip into the ws2815 strip to drive.
Win #2:
I scoped the data output of the ws2815 LEDs and the logic is at 5V.
Conclusion:
The IC is 5V logic.
P.S.
I wouldn't be surprised if trying to drive 12V logic would actually fry the chip. If somebody wants to test this that would be cool to know/understand.
I have a roll of ws2815 LED on my desk.
Win #1:
They do in fact run with 5V logic input. I actually just wired the GND/DAT from the end of a normal neopixel (ws2812) strip into the ws2815 strip to drive.
Win #2:
I scoped the data output of the ws2815 LEDs and the logic is at 5V.
Conclusion:
The IC is 5V logic.
P.S.
I wouldn't be surprised if trying to drive 12V logic would actually fry the chip. If somebody wants to test this that would be cool to know/understand.
NightRogue
New member
Well, I am not going to say that I "tested" this, but as a n00b messing around with this stuff, it took me frying multiple kits I put together for my Tesla before I was able to find the information I needed about the 5v data line. You can find this info NOWHERE it seems, and even searching for it was a chore until I knew exactly what the problem was.
Even with a 12v to 5v buck converter and using capacitors to even out power delivery, I still get issues with the first LED in each strip, and I can't seem to pump very much juice (brightness) into them before they start acting up. Flash, wig-out, etc. BUT, at least they still work and I haven't fried them. Probably something to do with the fact they are 144/m ws2815 running off one controller? I don't know I am using a signal amplifier on the data line and everything.
Anyhow, yeah you won't "FRY" the chip with 12v running through data right away, but after a few hours of running them, they will start to go on the fritz, which you can mitigate by dropping brightness for a little while, but eventually they will just stop responding to input.
- Status