Maker Pro
Maker Pro

EMI issue with USB connection

Hello there! I'm a hobbyist and I have a major and urgent issue.
I've made a USB device by using the PIC18F4550 and on the same board I have a power transformer (220 V -> 2x12V, 2A).
The problem is, whenever I turn the power transformer on/off (entering it into the mains outlet or decoupling it), it disconnects my USB device from the PC for 1 second and then it automatically reconnects. As I've read on the internet, it's an issue caused by EMI, but I don't know how to prevent it. The USB wire is a shielded B-Type...
And I'm sure the issue is caused by the transformer, cause even if it's not on the board, unconnected to anything, not even a rectifier bridge, at 20cm from the device, it disconnects my usb device...

Help please, I need to implement a solution very fast!

P.S. I'd rather not place the transformer in a separate box...
 
Try powering it from a d.c source like a battery or power supply. If it still does it then it's not the transformer or it's connections, lets rule that out first. Post any schematic and a PCB layout diagram also so we can see what we are dealing with.
Adam
 
Hi Istvan,

Please posting sharp picture(s) of your board as your schematic will help us to help you.
Olivier

Here are the images of the breadboard... I don't have a schematic yet for it, sorry... You can see that the transformer is unconnected, still, when I plug it in the mains, or pull it out of there, it briefly disconnects my USB connection for a second. Even if my laptop is running on battery mode, it does the same thing (There's no issue with the wall outlet).
 

Attachments

  • IMG_20140608_190945.jpg
    IMG_20140608_190945.jpg
    232 KB · Views: 138
  • IMG_20140608_191001.jpg
    IMG_20140608_191001.jpg
    198.5 KB · Views: 103
Try powering it from a d.c source like a battery or power supply. If it still does it then it's not the transformer or it's connections, lets rule that out first. Post any schematic and a PCB layout diagram also so we can see what we are dealing with.
Adam
I need to power it from a transformer, as I'll need it to power a solenoid later... DC power from a battery would waste fast.
 
Don't see the USB connection on your picture. If not yet the case unshielded wires must be very short between the PIC pins and the connector, around 1 inch maximum.
 
Don't see the USB connection on your picture. If not yet the case unshielded wires must be very short between the PIC pins and the connector, around 1 inch maximum.
Should I connect the shielding from the usb to the device's gnd?
 
The cable shielding is to be connected on the device GND.
The connector frame is to be connected to device GND via a ferrite bead or a 1-10nF capacitor, to be experimented
 
I need to power it from a transformer, as I'll need it to power a solenoid later... DC power from a battery would waste fast.

Yes I know I am not talking about replacing the transformer with a battery. I am talking about trying it to take out any problems that might be attributed to the transformer.
 
When it fails do you get a warning message from the PC or does it just disconnect and then re-connect?
That typical disconnected BAdumm sound (win 7), and reconnect sound (baDUMM) :) Nothing more. (Besides losing my dataconnection in my application, which is sensing atm the analog value of the phototransistors)
 
I have to go now. I'll give this issue some thought. If it was a switching regulator then I could understand but a linear transformer is not normally an issue. But it could be your wiring, trying to run a USB connection like this on bread board is challenging at the best of times. There are strict guidelines for USB and bread board is not one of them. :)
 
I have to go now. I'll give this issue some thought. If it was a switching regulator then I could understand but a linear transformer is not normally an issue. But it could be your wiring, trying to run a USB connection like this on bread board is challenging at the best of times. There are strick guidelines for USB and bread board is not one of them. :)
Probbly you're right, though I wouldn't want to make a pcb without the guarantee that it'll work...
 
Is there a way to make the induction-suction process at the transformer to slow down (resistors, caps?) when disconnected from powersource, so it won't make that much of a sudden noise?
 
Top