How to implement 10...
 
Notifications
Clear all

How to implement 100 null pixels?

68 Posts
11 Users
0 Reactions
10.8 K Views
Posts: 6728
(@p-short)
Illustrious Member
Joined: 19 years ago

They're only guaranteed to 4.1V (per the ATMega328P spec) with a 20mA load, although the output signals likely get fairly close to 5V under no load. In any case I was thinking of the Wemos D1 Mini, which does need an external level shifter (which could be an 75ACT126 powered by 5V).


Reply
Posts: 1420
Topic starter
(@1pet2-9)
Noble Member
Joined: 6 years ago

In practice, 4.1V is about what I get when I run a 5V Arduino off USB. If you're powering a buffer, I don't think you need to sink the full 20mA. You're talking a CMOS gate connection. Something like a D1 I would absolutely be using shifter (I'd be using my I2C shifters by now). I think the data line is driving basically a CMOS gate as well on the other end, but you are also driving the whole data run length for X number of meters.


Reply
Posts: 6728
(@p-short)
Illustrious Member
Joined: 19 years ago
1pet2_9 wrote:In practice, 4.1V is about what I get when I run a 5V Arduino off USB. If you're powering a buffer, I don't think you need to sink the full 20mA. You're talking a CMOS gate connection. Something like a D1 I would absolutely be using shifter (I'd be using my I2C shifters by now). I think the data line is driving basically a CMOS gate as well on the other end, but you are also driving the whole data run length for X number of meters.

The minimum Vih of the WS2811 is 0.7VCC = 3.5V according to the spec that I just looked at. The actual threshold is likely to be less than that, as 3.5V is what needs to be met in the worst case of voltage, temperature, process variations, etc. I would expect the output of a CMOS gate powered at 3.3V would work in most cases, but I wouldn't count on that for designing circuits.


Reply
Posts: 1420
Topic starter
(@1pet2-9)
Noble Member
Joined: 6 years ago

I mean the gate terminal of a CMOS (vs. the drain and source).


Reply
Posts: 105
(@pakeonoahu)
Estimable Member
Joined: 11 years ago
P. Short wrote:Per the spec, the WS2811 chips require >3.5V on their inputs to guarantee that they adequately sense the '1' signal. So generally the driver chip needs to be powered by 5V (not 3.3V).

The other thing to watch for is that the 74AC and 74HC chips have the same requirement on their inputs when they're powered by 5V to guarantee proper operation, so it's better to use the similar 74ACT/74HCT parts in their place. Neither the 3.3V nor the 5V Arduino's are guaranteed to drive their output signals high enough to guarantee proper 74AC/74HC operation as far as I can tell, so I would tell others to use 74ACT parts (the 74AC/74HC parts may appear to work, but there is no guarantee that this will continue over all temperature, power and noise conditions).

I think in my scenario was the NCP1117ST25T3G voltage regulator on the Nano, this regulator have a drop out voltage of 1.2v, I checked the voltage on the 5V pin out on the Nano when applying 5V to Vin, it show 4.1v and I think it's in the unstable/unknown state. Although the ATMega328P still functional, I put an O-scope on the 5V pin, DC looks very noisy, I also looked into SN74HC125N that I am using, it's HC( High-speed CMOS), datasheet show operating voltage range is 2-6V, 74xx series normally are TTL and operates on 5V, but I only see datasheet show low current on the lower operation voltage @ 20uA. I compared the WS2811 data stream on the output pin using O-scope when the 74125 was powered by Nano's 5V out vs 5V power supply. When it's powered by 5V power supply, Y output has 5v max on the pulses and I do see some spikes a little higher, does have some ringing but pixels has no issues, when powered by the 5V pin(Nano), when there's issue, the output pulses max was around 3V, some spikes are at 4v, waveform also have ringing signatures like using 5V power supply, pixels responds but with random colors, by power cycling the unit, sometime it does have full power 4.1V output and pixels acted normal, that's the reason I concluded the 5V regulator was operating at an unknow state and I shouldn't have ran it with 5V using Vin even when the ATMega328P looks fine.

ps, thanks for the info on ACT/ HCT series, I will keep that in mind when I need to refill next time.


Reply
Page 13 / 14
Share: