• Please review our updated Terms and Rules here

Smoothing a cheap universal power supply

Chuck(G) said
Do any of these things claim to comply with FCC Part 15, subpart B regulations?
Click to expand...
I'm getting the impression that the label on those things with all the pretty compliance icons is probably the most expensive component.

Wasn't there some noise, no pun intended, about micro-computers causing interference in TV's, that resulted in a whole bunch of new FCC guide lines ?
patscc
p.s. this is sarcasm, so no need to recount the early days of the personal computer
 
Chuck(G) said:
I wonder if the Canon PSU is a genuine Canon product or if it's a Chinese counterfeit knockoff.
Click to expand...

Definitely a knockoff. There is no way Canon would build a power supply like that -- it would never be able to pass FCC or UL certification.

The counterfeit power supplies often look identical to the real part, right down to all the text and symbols on the label, except that they weigh less:

 
And here is my own video demonstrating the massive amount of AM radio interference generated by cheap knockoff power supplies:

 
Heh, I just grabbed a Chinese PSU and I read the following on the label:

Some Chinese Guy said:
WARNING: (1) This device may not cause harmful interference and (2) This device must accept any interference received, including interference that may cause undesired operation.
Click to expand...

Say what? :huh:
 
With the AM test something in my room is causing a lot of interference, but it's not my real Canon CB-5L (DS8101), which came with my EOS 300D.

It also has the marking on the bottom stating "This device complies with part 15 of the FCC rules. Operation is subject to the two conditions:..." so at least this old Canon charger has FCC standards compliance.

My Real Nintendo DS charger however doesn't have any message stating that it complies with Part 15 of the FCC standards, however unplugging that didn't fix the interference, it's more than likely either my caseless desktop (what cases easily fit a 10.8" GPU?) or my laptop powercord, as the PA-10 isn't exactly designed with the best shielding on the cable (known for the IC that tells the laptop what cord it is becoming damaged from interference).
 
Compgeke said:
it's more than likely either my caseless desktop (what cases easily fit a 10.8" GPU?)
Click to expand...

Desktop PCs can and will generate a lot of radio interference with their case removed. Also that's why you're supposed to put metal covers over unoccupied expansion slots -- for RF shielding, as well as to help direct the airflow. Tightening the thumbscrews on cables (if any still have them!) helps, too.
 
If I may return to original topic....

Noting all comments above, I decided to do some more research and testing before taking chainsaw and arc welder to my cheap plastic PSU.

First, I noted that the RFI was audible but minor when the PSU was connected to my DIY adapter and to the ADC monitor. It became overwhelming when I touched the shield of the DVI lead (from the computer) to the shield of the DVI plug on the adapter. So massive RFI was being emitted from the ground plane and DVI cable shield.

I had checked the PSU output with a digital multimeter. That showed a stable 24.3vdc. Now I realised I needed to check output with oscilloscope. That revealed that the PSU output oscillates between +140v and -140v on a 20ms cycle - i.e 50Hz, which is AC mains frequency here. This pattern was on both positive (tip) and negative (sleeve) of the coax connector. There must be enough variance to produce the "average" +24.3vdc, but I couldn't see it.

I checked a couple of original maker's laptop PSUs and they did not show that radically oscillating output, just minor ripple. So I want to smooth out that 50Hz oscillation.

Here is the circuit layout for the adapter that receives the output from the PSU:

ADC-DVI_Wiring-300x267.jpg

I still seek opinions on whether a capacitor between the positive and ground power rails on the adapter card can smooth the current, at least to a reasonable degree. If so, what value?

Still hoping I don't have to build a linear PSU from components costing more than my salvaged monitor!

Rick
 
When using a DMM, ripple can increase if a PSU is not under a proper load. Where you testing with the PSU while the monitor was hooked up ?
You might want to try a basic RLC smoothing network, something like:

---R---L----
| |
| |
C1 C2
| |
| |
---------------

The ASCII schematic is kinda crude, but hopefully you get the idea.
patscc
 
patscc said:
When using a DMM, ripple can increase if a PSU is not under a proper load. Where you testing with the PSU while the monitor was hooked up ?
Click to expand...

You are right - the ripple is much more normal when the full load is in place. But still significantly greater than the other PSUs compared.

You might want to try a basic RLC smoothing network, something like:

---R---L----
| |
| |
C1 C2
| |
| |
---------------

The ASCII schematic is kinda crude, but hopefully you get the idea.
patscc
Click to expand...

I had a look at some circuits for smoothing networks. A bit beyond my competence :-?

In the end I decided to tap the mat on this one. I bought a 24vdc regulated wall-wart for not much more than the components cost, and less hassle. It eliminates all RFI and is fit for purpose.

I note that the cheapy PSU is good value ONLY in the earth-isolated situations for which it was designed. As a laptop supply with no earth through the mains plug, the RFI is tolerable. When it is connected to an earthed system, the cable shielding and earthed chassis becomes a high-gain antenna for the PSU ripple in the return circuit, and RFI goes through the roof.

Another learning experience - thanks all for inputs.

Rick
 
You must log in or register to reply here.
Back
Top