1W
https://it.aliexpress.com/item/1005006216595025.html
2W
https://it.aliexpress.com/item/1005006237065080.html
These should be equivalent. But there is a problem: the 12V version accept only 11.5 to 12.6V in. I range from 10.5 to 14.6V.
I would like to avoid a chain like non-isolated DC DC to 12V then isolated DC DC to 5V. At that point, I wire the Apple charger and bye bye direct battery PSU.
Can you suggest a wider Vin? Also a higher power module (to keep as test for other purposes)? Thanks
Last edited:
The nearest is 2m the farest is 7m.
Since sensor are hidden in the van walls I can't. But hey, with powerbank it works flawlessy. I suspect that I have problems on LIFEPO4 lines.
230VAC > 12VDC > LM317 DeNoiser > Arduino (random reads fails)
Lifepo4 12.8V > LM317 DeNoiser > Arduino (no reads at all!!)
That justify also the problems with LM317 NORMAL module, DCDC 2x USB A, and the buck/boost DCDC I bought for (called DcDc in these posts).
Ive got VICTRON MPPT all time conected, each battery has a bms with BT module, but nothing more...
I will buy a Isolated PSU and I will check.
Meanwhile I will clone that NANO and work on bench to record these images with my scope.
Im not sure to have correctly understand. I power the DHT with 5v; then between + and DATA I put a 10k ohm resistor on each sensor.
Posting datasheets always helps.
https://cdn.sparkfun.com/assets/f/7/d/9/c/DHT22.pdf
Page 5 -
6. Operating specifications:
(1) Power and Pins Power's voltage should be 3.3-6V DC. When power is supplied to sensor, don't send any instruction to the sensor within one second to pass unstable status. One capacitor valued 100nF can be added between VDD and GND for wave filtering.
Ive read the datasheet.
Vcc is 5V.
I dont understand the second sentence. I suppose you shouldnt ask too frequently (<1s) a read to avoid unstable status.
I guess the 100nF should be put as near as possible to the sensor.. I can't do that anymore.
You can make stronger pullup from arduino side, just add a resistor 10k from one of sensors data line to +5V . And check if that has any effect.
Powering from completely isolated source like battery is anothet thing, then all potentials, signals and grounds, float, and there is no induced noise, because same noise is induced in all wires i think. Entire arduino and sensors float and maybe that helps overcome noise problem, if that's noise fault. Isolated converter can have capacitors from output to input, not all are completely isolated. In DC terms there's insulation, but input to output has some capacity, and switching noise from primary winding may pass output. If there's a capacitor between input/output, they return that leakage noise to input, but at cost of refuced insulation.
Powering from completely isolated source like battery is anothet thing, then all potentials, signals and grounds, float, and there is no induced noise, because same noise is induced in all wires i think. Entire arduino and sensors float and maybe that helps overcome noise problem, if that's noise fault. Isolated converter can have capacitors from output to input, not all are completely isolated. In DC terms there's insulation, but input to output has some capacity, and switching noise from primary winding may pass output. If there's a capacitor between input/output, they return that leakage noise to input, but at cost of refuced insulation.
Also, using Vin pin of Arduino board has some limitations, voltage must not exceed 12v if i remember right, for AMS1117 regulator.
Are the input pins of your arduino 5V tolerant? On battery is VCC 3.3V and on smps 5V into the arduino? Is the DHT also driven by VCC (3.3V battery mode, 5V smps)?
Ive found this one.
Isolated DC to 5V USB A 1A 5W
https://it.aliexpress.com/item/1005001668186115.html
(Seems a classic flyback isolated octo coupled circuit. The best isolation, right? )
Alongside with
Because Vin uses ARDUINO 1117 on board regulator that is worse than LM317 DeNoiser to achieve 5V DC.
It can go up to 15V and even more but all relay on 1117 5V regulator.
Feed a proper 5V like usbC or other sources is better (if yoi dont swap - as said - the 1117 with a better version).
Arduino is 5V GPIO
RASP PI is 3.3V GPIO
My source derive from 230v (phone charger)/12.8v (lifepo4)/3.7v (lipo-liion battery powerbank) but every one become 5V DC.
Last edited:
Ok. I've found the issue. The inverter. When I turn on the inverter, it spread noise across input 12V connectors and then to every component connected.
No problem at all with everything (TUYA light led strip and TUYA 10X relay) except the Arduino. But at least I know the issue now.
So with inverter turn off:
Since these sensors are useful to check water infliltration and humidity checks, this compromise (inverter off while checking these datas) is good.
I can implement a relay to switch between dc and ac when inverter is turned on so I will have datas in both situation but that is not necessary.
It is more interesting to filter that dirty inverter-induced dc to let arduino works while inverter is on.
I will analyse it with DSO and upload results here.
If you can suggest me how to proper analyze it, tell me 🙂
No problem at all with everything (TUYA light led strip and TUYA 10X relay) except the Arduino. But at least I know the issue now.
So with inverter turn off:
- Vin 13.6V direct feed Arduino works
- L7805 works
- DeNoiser LM317 works
- DCDC 5V USB-A works
- etc...
Since these sensors are useful to check water infliltration and humidity checks, this compromise (inverter off while checking these datas) is good.
I can implement a relay to switch between dc and ac when inverter is turned on so I will have datas in both situation but that is not necessary.
It is more interesting to filter that dirty inverter-induced dc to let arduino works while inverter is on.
I will analyse it with DSO and upload results here.
If you can suggest me how to proper analyze it, tell me 🙂
Ok... Inverter... And maybe not pure sinewave ,but "modified sinewave" , which is actually a squarewave ?
Im noob but not so dumb xD It's a PURE SINE WAVE 2500W that I use for fridge, air conditioning, cooking, cleaning and PC gaming.
Meanwhile I burnt the NANO....
After connecting to 12V Vin I tried other method. While performing one, "TUYA 10x channel relays" exctraction of 5V DC from USBC port (soldered wires to + and GND from board) I just connected + to arduino 5v and GND was on 12v GND.
Well, TUYA has 0.3V between USBC GND and supply GND...
SO I powered it with 5.6V since USBC was 5.3V and GND difference 0.3V.
Max allowed from CPU is 5.5V. Three holes in the chip. 🥲
If inverter is pure sine wave advertised one , then probably it makes noise not at 220V output ,but at 12V input wiring , and you may try to put ferrite beads for EMI filtering , and capacitors between + and - input wires ,also try same bead on AC output cable ,maybe it will radiate less . About arduino - i doubt if chip would explode by 0,3V overvoltage ,also voltage drop on socket and wires ... probably it was small spark while connecting power ,or power voltage was 12v , and some input protection diodes burned /shorted ,and ten chip burned . In similar way i have burned several arduino pro mini boards ,whose have no usb socket ,but have 6 pins ,and need to use external board with usb socket and converter chip CH340 on it . When data pins connected first ,when some of those 6 pins lost contact ,all power is going through data wires and diodes inside chip ,and diodes are weak ,and shorted ,chip becomes hot ...Maybe thats your issue too . From then , i always first connect programming board to arduino pro mini with proper contact ,and only then plug to pc usb socket .
If the input to the inverter is 12V and that 12V is grounded to the camper shell, yikes. 2500W is over 200A@12V. There is going to be some noise on the ground line!
Input is 12.8V lifepo4 but negative is not grounded.
Earth of inverter (that it is fake, it just connect inverter chassis to shuko socket earth) is connected to camper shell (that is a van, so il full metal) and other components' chassis (like MPPT).
Also the engine battery (obv) is grounded to shell, and it is connected to isolated DcDc to recharge Lifepo4 from engine alternator.
Exactly my point of view.
In fact, when inverter is on, without DeNoiser, the NANO even works constantly and display power on and off. With DeNoiser and other DCDC the NANO light up the display, sometimes read datas, but don't work properly.
I'll try beads, thanks!
Last edited:
https://electronics.stackexchange.com/questions/283228/how-to-reduce-noise-caused-by-inverter
That's the same problem similar setup.
I've got Shunt resistor (victron) too.
Tomorrow I'll make the same measurements and we will compare them.
At least I know the problem.
But my inverter wires are 70mmq so is better to work only on the source needed, the tiny 1.5mmq wires that supply current to NANO and other low consumption (<60W) devices.
AmyAlice now seems necessary.
When I worked on sensitive industrial electric cabinet, I saw this FN2090-10-06 (the img found is similar but for AC. I can't find schematic of DC model).
That's the same problem similar setup.
I've got Shunt resistor (victron) too.
Tomorrow I'll make the same measurements and we will compare them.
At least I know the problem.
But my inverter wires are 70mmq so is better to work only on the source needed, the tiny 1.5mmq wires that supply current to NANO and other low consumption (<60W) devices.
AmyAlice now seems necessary.
When I worked on sensitive industrial electric cabinet, I saw this FN2090-10-06 (the img found is similar but for AC. I can't find schematic of DC model).
Last edited:
My INVERTER ON:
Inverter off (same V scale):
Inverter off (100mV scale):
NB: when inverter is off, MPPT, tuya relays, router wifi, victron shunt are on. For that in 100mV range we see spikes.
If I see right when inverter is ON it generates spikes every 20 micro seconds (50kHz) that ranges up to 5.3V pkpk!!!
Fortunatly other DC components seems not affected by that.
If this kind of disturb could interfere with something tell me what could be.. ex. Arduino nano Vin 😉
Here the AC OUTPUT INVERTER analysis:
Thank everyone, all always
Last edited: