Hy,
Long story short, I'm making android tablet car head unit, I removed the battery and I'm powering it with 4.2v via regulator (to the battery connections).
The tablet mcu is generating digital (dial up modem like noises) into the power feed, based on the tablets activity.
LM317K nor the LM2596S will filter this out, so I'm guessing the noises are not on the positive feed, but on the ground.
Any audio amplifier I connect to the same power supply will pick up these noises, as system will be installed in the car, I cannot feed this to cars ECUs.
I need it gone.... This project was halted 2 years back and now I'm trying to finaly solve the issue.
Also any aditional current must be minimal as it will be on all the time (even when car is off).
Only tablet is connected via regulator any I can see this noise on both sides of the regulator:
Please shine a light on me, this project would be considered finished if it wasn't for that noise.
Long story short, I'm making android tablet car head unit, I removed the battery and I'm powering it with 4.2v via regulator (to the battery connections).
The tablet mcu is generating digital (dial up modem like noises) into the power feed, based on the tablets activity.
LM317K nor the LM2596S will filter this out, so I'm guessing the noises are not on the positive feed, but on the ground.
Any audio amplifier I connect to the same power supply will pick up these noises, as system will be installed in the car, I cannot feed this to cars ECUs.
I need it gone.... This project was halted 2 years back and now I'm trying to finaly solve the issue.
Also any aditional current must be minimal as it will be on all the time (even when car is off).
Only tablet is connected via regulator any I can see this noise on both sides of the regulator:
Please shine a light on me, this project would be considered finished if it wasn't for that noise.
For the tablet I had made custom launcher, adjusted resolution, made a frame from steel, glued the tablet to the frame and frame to the console in car, I had soldered all the wires for buttons, usb, audio to it's motherboard, ruined 2 motherboards in the process, made a custom turn-on animation.... dab+ and so on and so on... changing the tablet is not an option...
Seems like a crazy idea for something that quickly ages tehnologicaly, but I am only using the poweramp player and dab+ app, so I'm not concerned about that.
Basicly it is a big project, but I had removed everything besides the tablet and the regulator for testing... there is also arduino, relays, dab+ usb stick, usb hub, ground loop isolator, 4ch amp....
Attachments
Have you tried a common mode choke?
https://www.coilcraft.com/en-us/edu/series/a-guide-to-understanding-common-mode-chokes/
https://www.coilcraft.com/en-us/edu/series/a-guide-to-understanding-common-mode-chokes/
I had try to make one, taking the turns of wire off, few by few and the noise is actually louder with it.
The tablet is changing current alot and fast. I can't even turn it on with 10r resistor in series, voltage starts oscilating with 1v diff.
Hy, yes I have seen the mail earlier at work, I had just arrived home and had stopped at the store, but sadly the don't have any isolated converters in stock.
If possible I would like to solve the problem with common mode choke like dreamth suggested, or some other filtering method, just because I need to keep current as minimal as possible, but I have a feeling it wont do good.
What makes you think that way? The fact that common mode choke makes the noise louder or the fact that voltage is oscillating when I use resistor in series?
Voltage starts oscillating after the resistor (between the resistor and the tablet), voltage remains stable on the side between the resistor and the regulator.
It is doing the same with SMPS, linear or when connected directly to the bench PSU.
As far as my knowledge goes I cannot think of a reason for this behaviour, other than tablet is changing it's resistance way too much and because "U = R x I properties", voltage goes crazy.
I had tryed various capacitors on the tablet side, but it just changes the frequency of the oscillation.
Lets say bench PSU adjusted to 4.2v directly to the tablet. This is the circuit with minimal components / wiring and problem remains. When I add resistor in series on V+, voltage between the resistor and the tablet is oscillating.
Maybe try a couple isolation transformers to keep the tablet ground and the vehicle ground isolated when hooked up? Then try to see if a commercial common mode choke will work.
I had a tube preamp running off of a SMPS a while back and this was the only way the noise was dealt with.
I had a tube preamp running off of a SMPS a while back and this was the only way the noise was dealt with.
1st reason is the heat. I'm rebuilding 1991 camaro and it would be the shame if it would get burned because of the chinesse tablet battery.
2nd reason is that I'm using usb host and while in host mode I cannot charge it...
I'm on to something. 3x 4700uF caps almost eliminate the noise... adding another one don't change anything... will try to add some ceramic low value...
You think that I can wire together more caps, add diode to prevent backfeeding the regulator, maybe voltage wont oscilate and that would allow me to add 100ohm resistor to minimize the noise further and call it a day?
EDIT:
As the afterthought no... current spike would probably destroy the smps while charging the caps....