Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Test Bench : SMPS240QR from Connexelectronics
Test Bench : SMPS240QR from Connexelectronics
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 14th January 2017, 03:17 PM   #1
peufeu is offline peufeu  France
diyAudio Member
 
Join Date: Mar 2001
Location: Lyon, France
Default Test Bench : SMPS240QR from Connexelectronics

This will be a test bench report of Connexelectronics SMPS240QR, +/- 30V version. I won't copypaste the specs from the PDF linked above, only the important matters.

Input : 230VAC
Outputs : +/- 30V and +12V
Topology : quasi-resonant, rated for 240W output on the +/-30 rails.
On-board Mains filter.
Unregulated output. This means the output follows the voltage on the mains input cap.
No Power Factor Correction.
Reasonable price (55).

Initial impressions :

Positive :
- Small, compact, light.
- The supply seems well built, PCB is a real double sided FR4 PCB
- There is a temperature sensor glued to the switching transistors
- Caps are 105C... all in all, it looks good.

Negative :
- Schematics are unavailable.
- Layout is a bit tight around the output connector and requires nimble fingers
- Resonant capacitor looks a bit small.

Necessary parts not included in the package :
- Insulating sheet to be placed under the board, if mounted on a metal plate. I used photocopier-safe transparency film, which is quite thick and resists high temperatures.
- Silpads to be placed between the bottom-side secondary diodes and the metal enclosure.

Tests under load

Code:
I(V+)   V+ (V)      Ripple  I(V-)   V- (V)     Ripple
(Amps)  (Volts)     (Vpp)   (Amps)  (Volts)    (Vpp)
0       +32.3       --      0       -32.4       --
0       +31.7       --      2       -30.7       0.640
2       +30.6       0.680   0       -33.6       --
2       +29.9       0.980   2       -29.9       0.940
3.7     +28         2.2     3.7     -28         2.2
Well, it works and it is within spec*, so no problem here. Considering it is unregulated, it could be interesting to invert one of the amp channels to spread the load between both polarities.

* = Ripple would be out of spec if this was the 32V version, but this is the 30V version, which qualifies as "custom" and has relaxed ripple requirements... okay, this might be cheating a little bit...

Attached files :
- ripple with 2A load on V+, V-, and both.
- ripple with 3.7A load on both outputs.

Next : EMI/RFI/Noise
Attached Images
File Type: png 01 ripple_v-_2A.png (11.7 KB, 278 views)
File Type: png 01 ripple_v+_2A.png (9.4 KB, 277 views)
File Type: png 01 ripple_both_2A.png (10.9 KB, 276 views)
File Type: png 01 ripple_both_3.7A.png (14.8 KB, 273 views)

Last edited by peufeu; 14th January 2017 at 03:41 PM.
  Reply With Quote
Old 14th January 2017, 03:23 PM   #2
ravels is offline ravels
diyAudio Member
 
Join Date: Oct 2016
What happens at 3A or 4A load?
  Reply With Quote
Old 14th January 2017, 03:44 PM   #3
peufeu is offline peufeu  France
diyAudio Member
 
Join Date: Mar 2001
Location: Lyon, France
I added 3.7A load, which is the best I can do with these huge power resistors. I could add the extra resistors to reach 4 amps, but that would require locating some more power resistors... and I'm not really motivated.

If you guys know what the display usually looks like on Rigol DS2000 scopes, you're probably wondering why I have hires/time averaging mode turned on. Well, there was a bit of a snag about the noise...

Part 2 : EMI/RFI

Measuring the output voltage using the traditional 6" scope probe ground alligator clip wire is a futile effort, as is usually the case when in the vicinity of a SMPS. I will not post the scope shots, because these measurements are all fake, since the ground alligator clip wire behaves as a nice loop antenna of unspecified characteristics.

Now, what would happen to, say, an audiophile circuit located a few inches away from this power supply? Some magnetic coupling will occur, so I will explore around the supply using a simple loop antenna, loaded with 50R and fed into the scope with a coax. Flattening the loop discriminates between capacitive and inductive coupling. If the squeezed flat loop still picks it up, it's capacitive. If only the nice round loop picks it up, it's inductive.

So far, the transformer seems to leak quite a bit of magnetic field. As expected, worst spots are above and below, then to the sides (along the length of the board).

The front and back of the board (where the connectors are located) are much quieter.

Scope shot : voltage from aforementioned loop antenna, placed as shown on photograph.
Attached Images
File Type: jpg 02 loop.jpg (176.2 KB, 266 views)
File Type: jpg 02 flat loop.jpg (170.8 KB, 82 views)
File Type: png 02 induction loop 2.png (13.2 KB, 79 views)

Last edited by peufeu; 14th January 2017 at 03:58 PM.
  Reply With Quote
Old 14th January 2017, 03:47 PM   #4
KSTR is offline KSTR  Germany
diyAudio Member
 
KSTR's Avatar
 
Join Date: Jul 2007
Location: Central Berlin, Germany
It's a pity that the output scales with mains (no universal mains input), because that's one of the main motivations to use SMPS for me, besides form factor, efficiency etc.
Also, the single +12V/0.3A Aux ouput is a bit awkward for a front end supply in an amp or active speaker etc.

Still a nice product, and thanks for the test.
  Reply With Quote
Old 14th January 2017, 03:59 PM   #5
yu3ma is offline yu3ma  Serbia
diyAudio Member
 
yu3ma's Avatar
 
Join Date: Jul 2014
Location: Kragujevac, Serbia
@peufeu

On my FSFR2100SX LLC converter output ripple looks like this (no averaging, only BW limited channel, same DS2000 scope):

Test Bench : SMPS240QR from Connexelectronics-ds2_quickprint42-png

It is noticeable almost pure sinusoidal ripple and very small glitches produced in dead-time period of switching sequence.
Attached Images
File Type: png DS2_QuickPrint42.png (39.2 KB, 284 views)
__________________
DIY electronic projects - Forum of applied electronics
http://forum.yu3ma.net/index.php
  Reply With Quote
Old 14th January 2017, 04:08 PM   #6
KSTR is offline KSTR  Germany
diyAudio Member
 
KSTR's Avatar
 
Join Date: Jul 2007
Location: Central Berlin, Germany
Found one SMPS at connex that looks nice with universal mains input, the SMPS250QR, but alas no datasheet. Vaporware? Same is true for many other models they offer, makes me puzzled.
  Reply With Quote
Old 14th January 2017, 04:16 PM   #7
peufeu is offline peufeu  France
diyAudio Member
 
Join Date: Mar 2001
Location: Lyon, France
It would be better if it was regulated, yeah...

But resonant and quasi-resonant supplies don't have enough range for a universal input (100-240V) anyway. An active boost PFC would be needed to achieve this, adding to cost.

yu3ma : Your output is very clean. How is your measurement setup? (cables, load, etc?)

Cheap EMI tweak #1 : copper tape

I decided to simulate the layout I had in mind for my amplifier enclosure, which is the usual model with heat sinks on the side. I would have paced the supply in the middle, mains side facing the back panel, and one amp board on each heatsink, on both sides of the supply.

So, I set the loop antenna's edge 5cm away from the edge of the supply PCB, and measured induced noise, which was about 2.8 mV RMS into 50R.

I then added a copper band around the transformer (see photo), to create a shorted turn around the whole transformer (not around the secondary!). This reduces flux leakage, as shown by before/after scope shots.

Noise is reduced about 18dB by this very simple tweak! Why the manufacturer did not include this is beyond me.

I should probably connect the copper tape to Earth or primary side GND to reduce capacitive coupling between transformer windings, though.
Attached Images
File Type: jpg 03 copper 3m.jpg (184.6 KB, 106 views)
File Type: png 03 shortingring_none.png (15.8 KB, 104 views)
File Type: png 03 shortingring_3M.png (15.8 KB, 46 views)
  Reply With Quote
Old 14th January 2017, 04:16 PM   #8
yu3ma is offline yu3ma  Serbia
diyAudio Member
 
yu3ma's Avatar
 
Join Date: Jul 2014
Location: Kragujevac, Serbia
@KSTR
PFC stage can be quite polluting, it produce much more RFI/EMI noise than converter itself (especially if we compare to pure LLC converter) because in many cases PFC works in hard-switching regime.
__________________
DIY electronic projects - Forum of applied electronics
http://forum.yu3ma.net/index.php
  Reply With Quote
Old 14th January 2017, 04:25 PM   #9
peufeu is offline peufeu  France
diyAudio Member
 
Join Date: Mar 2001
Location: Lyon, France
EMI tweak #2: Canned vegetables

The picture explains it all (along with showing the position of the measurement loop).

The copper tape is still there. I had enough trouble sticking it in to NOT want to tear it out just to have more dynamic range for the shield measurement. Therefore, the amount of noise to be measured is quite small, and scope noise becomes a major factor.

Besides, we're interested in the sawtooth signal (which is inductively coupled), not by the switching spikes (which are coupled by some other means), and these spikes represent a non-negligible proportion of the measured signal. Therefors, the scope's displayed RMS values should not be used.

So, I will instead eyeball the sawtooth signal and say the shield brings something like a 10-20dB improvement. Which, again, isn't that bad for a 0.50 can of cooked vegetables.

Advice for Connexelectronics: Make a sheet metal shield enclosure for a cost of $0.5 ; sell it along with the PSU for $5. No problem here.

Anyway. With the shield, I'm gonna say the noise is low enough to stick it inside the amp's enclosure, but I'll rotate it 90 and mount it as close as possible to the front panel, to keep it away from the inputs etc, which are of course on the back panel.
Attached Images
File Type: png 05 smallshield off.png (15.9 KB, 42 views)
File Type: png 05 smallshield on.png (15.6 KB, 24 views)
File Type: jpg 04 shield can small.jpg (190.6 KB, 82 views)
  Reply With Quote
Old 14th January 2017, 04:26 PM   #10
sdman is offline sdman
diyAudio Member
 
Join Date: Oct 2001
Location: Southwest
Default Test Bench : SMPS240QR from Connexelectronics

Peufeu,

Rather clever adding the copper band around the xfm.

Sam
__________________
Sam
  Reply With Quote

Reply


Test Bench : SMPS240QR from ConnexelectronicsHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
test bench lighting lmartine3 Equipment & Tools 48 4th January 2010 07:34 AM
Bench test KEEBIE Car Audio 4 1st November 2006 04:11 AM
Speaker Bench Test evilorange Multi-Way 9 13th October 2005 03:30 PM
What Test Equipment Do You Have On Your Bench?? ultra-lo-fi Everything Else 56 14th June 2005 02:47 PM
ultimate test bench camusmuse Everything Else 2 13th September 2004 06:59 PM


New To Site? Need Help?

All times are GMT. The time now is 03:05 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki