Author Topic: Full bridge rectifier shows wrong curve on scope  (Read 5587 times)

0 Members and 1 Guest are viewing this topic.

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Full bridge rectifier shows wrong curve on scope
« Reply #25 on: January 06, 2020, 04:02:47 pm »
"One probe is on the positive bridge rectifier output (red/orange wire), i.e., scope channel 3 (pout). The negative output (black wire) is connected to scope channel 4 (nout)."

At least the overall shape of those 2 bottom traces look close enough to the bottom 2 on my simulation, and would be even closer if the screen's 0V were in the same place.:)

"and checked it was really going from 1.65V to -1.65 V (approximately)."

Barely enough voltage for a bridge rectifier.

I don't know why the scope's ( CH3-CH4 ) is not working, it looks like it's doing ( Noise-CH4 ).

It looks like you might have one probe on X1 and one on X10, which would help to confuse the MATH.
.  That took much longer than I thought it would.
 
The following users thanked this post: petert

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Full bridge rectifier shows wrong curve on scope
« Reply #26 on: January 07, 2020, 01:51:36 am »
Your math is trying its best..... even though "it's cheating".   >:D

But what do you expect the math to look like when one of your signal channels is ten times the amplitude of the other?

I'll put this here again so people don't have to look for it on the first page.  Consider this an exercise in scoposcopy.    :wtf:



(please note that all the numbers have meaning and correspond to what is actually shown by the traces on the graticule)
« Last Edit: January 07, 2020, 01:57:55 am by alsetalokin4017 »
The easiest person to fool is yourself. -- Richard Feynman
 
The following users thanked this post: petert

Offline petertTopic starter

  • Regular Contributor
  • *
  • Posts: 178
  • Country: de
Re: Full bridge rectifier shows wrong curve on scope
« Reply #27 on: January 07, 2020, 10:15:30 am »
Thanks, you were both (@StillTrying, @alsetalokin4017) right regarding the scaling of the probes. I was trying out so many things I didn't notice that anymore haha.

After I had both probes on the same setting, it got a similar result to yours, even if somewhat noisy.

Regarding the "cheating": the reason was that I was trying to find out why the isolation does not work after unplugging the earth lead from both the scope and signal generator, which still puzzles me. (Which was a good example use of the math function, though ;)


Maybe weird house wiring? A UPS seems like the easiest way to isolate the signal generator (if I had one :D). Any other ideas?
« Last Edit: January 07, 2020, 10:25:09 am by petert »
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Full bridge rectifier shows wrong curve on scope
« Reply #28 on: January 07, 2020, 11:35:51 am »
Marvelous, I looked here to see if you'd got there yet. :)

I'm not 100% convinced that your generator's 0V is earth grounded, can't you measure Ohms between the scope's GND clip and the generator's 0V, anything less than about 1k Ohms would mean it is grounded.

"somewhat noisy."

A BR's DC output is often a bit noisy without a few k of load on the DC side to keep the diodes conducting with no non-conducting open circuit time.

Even when the input side IS floating you still can't short the input to output of a BR with 2 scope probe GNDs. To capture this one I saved the input sine on screen and then moved the scope's GND and probe to the DC output side.
https://www.eevblog.com/forum/beginners/fwb-full-wave-bridge-rectify/msg1117268/#msg1117268


.  That took much longer than I thought it would.
 
The following users thanked this post: petert

Offline Ysjoelfir

  • Frequent Contributor
  • **
  • Posts: 543
  • Country: de
Re: Full bridge rectifier shows wrong curve on scope
« Reply #29 on: January 07, 2020, 02:02:49 pm »
Just a shot in the blue: To isolate the grounds of your sig gen and scope... you didn't happen to use a multi-outlet extension cord, isolated the ground of that, and plugged both devices into the same extension? That would make both devices basically isolated from mains earth/ground, but the grounds of the scope would still be connected to the ground of the signal generator.



This whole "isolation"-Thing was a perfect example that university professors are sometimes dumb as hell, as one of those blew up three brand new scopes before he trusted me when I told him that he is right now shorting the 3 phase mains over the gnd clip... I got a bad grade in the lab report for pointing that out though...
Greetings, Kai \ Ysjoelfir
 
The following users thanked this post: petert

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: ca
Re: Full bridge rectifier shows wrong curve on scope
« Reply #30 on: January 07, 2020, 04:09:50 pm »
I did the same experiment and mine does the same thing. It's like the small voltage from the generator doesn't go below zero.So it's not a true AC sine wave.Simplest solution is use a low voltage transformer to test your rectifier.

That's a pretty terrible signal generator if the output can not swing below ground.  Easiest method would be to add a capacitor in series with the sig gen output to remove the DC offset.

It's an Atten ATF20B .It was all I could afford . Still not sure of many of the functions and how to use it. I'll figure it out eventually.
 

Offline petertTopic starter

  • Regular Contributor
  • *
  • Posts: 178
  • Country: de
Re: Full bridge rectifier shows wrong curve on scope
« Reply #31 on: January 07, 2020, 06:40:02 pm »
Just a shot in the blue: To isolate the grounds of your sig gen and scope... you didn't happen to use a multi-outlet extension cord, isolated the ground of that, and plugged both devices into the same extension? That would make both devices basically isolated from mains earth/ground, but the grounds of the scope would still be connected to the ground of the signal generator.
Well, that's embarrassing. It turns out you were guessing right! I didn't think of the "multi-socket shorting independent of house earth". But I could have sworn I tried all possible combinations (isolating the scope plug, isolating the signal gen plug, isolating the plug of the multi-outlet extension cord / multi-outlet power strip), so it should have worked anyways.

Maybe I had the right isolation (by "chance"), but the signal didn't look clean enough for some other reason and that threw me off course...
But it's good I made the mistake to be aware of my misconception about the multi-outlet power strip.

Thanks! Better knowing why you are a fool than staying wrong while clueless haha (sorry about the professor).

I was a bit nervous about testing the power plugs with the multi-meter, but I suppose I could have done continuity testing between the scope bnc and signal generator bnc plugs (the outer shells), safely.

Doing that indeed confirmed the (now obvious) direct connection between the earth plugs in the multi-outlet power strip (and therefore the scope and signal gen). I somehow assumed that the connection would just be made in the house wiring, later (and therefore not occur if the main's earth is not connected). But since the multi-outlet strip gets connected to the wall outlet by a single power plug, it should have been obvious that this could not be the case (and all individual sockets earth must be joined together). Oops.

I'll post a "success scope" image and a summary of the essential points, in case someone else stumbles upon this again.

Thanks again everyone for your patience.
« Last Edit: January 07, 2020, 06:52:10 pm by petert »
 

Offline Mr. Scram

  • Super Contributor
  • ***
  • Posts: 9821
  • Country: 00
  • Display aficionado
Re: Full bridge rectifier shows wrong curve on scope
« Reply #32 on: January 07, 2020, 10:43:43 pm »
I actually typed that question and solution a couple of days ago, then deleted it because you mentioned you isolated the signal generator after that. Should have gone with my hunch.
« Last Edit: January 07, 2020, 10:46:39 pm by Mr. Scram »
 

Offline Vovk_Z

  • Super Contributor
  • ***
  • Posts: 1478
  • Country: ua
Re: Full bridge rectifier shows wrong curve on scope
« Reply #33 on: January 07, 2020, 11:54:51 pm »
It looks like the bottom-right diode is in the wrong row ?
+1. It's an answer.
 

Offline petertTopic starter

  • Regular Contributor
  • *
  • Posts: 178
  • Country: de
Re: Full bridge rectifier shows wrong curve on scope
« Reply #34 on: January 08, 2020, 12:13:22 am »
For reference, a related two part video, which talks about full bridge rectifiers and "Oscilloscope Earthing / Isolation Challenges":

 

Offline petertTopic starter

  • Regular Contributor
  • *
  • Posts: 178
  • Country: de
Re: Full bridge rectifier shows wrong curve on scope
« Reply #35 on: January 08, 2020, 12:31:26 am »
Summary/solution:

As mentioned earlier, the breadboard circuit was correct, and the isolation was the main issue (multi-socket power outlet connects earth pins, even when the multi-socket as a whole is not connected to household's main earth).

Using a 10k load resistor in the bridge rectifier gave the best result, regarding noise and amplitude. Less creates some clipping, more creates distortion without increasing the amplitude. Using a probe with a 1X impedance setting (with the signal generator being isolated from earth ground/the scope's ground) instead of 10x, does not have any effect on distortion (contrary to what viewer comments claimed in the videos I linked right above this post).

Bridge rectifier output with 1K load resistor (and 10Vpp 50Hz sine input):

Bridge rectifier output with 10K load resistor (and 10Vpp 50Hz sine input):


Alternatives to isolating the signal generator's (or the scope's, or both) earth pin from the household's main earth are a UPS (uninterruptible power supply) or an isolation transformer. The former usually runs for some minutes only, the latter will still circumvent the RCD (that protects from dangerous currents through your body).

For differential probing using the math function: make sure both probes are set to the same impedance, and ensure the scaling on the scope (the software side) for the channels match the impedance of the attached probes as well.
« Last Edit: January 08, 2020, 12:50:27 pm by petert »
 

Offline Ysjoelfir

  • Frequent Contributor
  • **
  • Posts: 543
  • Country: de
Re: Full bridge rectifier shows wrong curve on scope
« Reply #36 on: January 08, 2020, 11:55:05 am »
Glad to hear that you solved the problem :)
Greetings, Kai \ Ysjoelfir
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Full bridge rectifier shows wrong curve on scope
« Reply #37 on: January 08, 2020, 11:34:30 pm »
 :-+ :-+

Sneaky ground loops can be really puzzling. I have an old analog FG (Interstate/Racal F43) which has a very convenient feature: a chassis isolation switch. So it can be "normally" connected through the line cord to mains ground, or floated by simply switching the switch. But ground loops are sneaky: if you connect, say, a BNC patch cord from the isolated FG to the non-isolated oscilloscope so you can monitor the FG's output ... guess what.
The easiest person to fool is yourself. -- Richard Feynman
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Full bridge rectifier shows wrong curve on scope
« Reply #38 on: January 09, 2020, 01:39:21 am »
... guess what.

The BNC socket is connected to the chassis ? :-//
.  That took much longer than I thought it would.
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Full bridge rectifier shows wrong curve on scope
« Reply #39 on: January 09, 2020, 11:06:28 am »
Ah... yes. Or no, depending on which BNC connector and which chassis you are referring to.

That is, on this FG, the power supply section _is_ connected normally, that is, the mains line ground is connected to the chassis. But the signal output section sends its output to the world through insulated BNC feedthru connectors that are _not_ electrically connected to the chassis at all.... unless the FG's isolation switch is in the normal "grounded" position.  To keep idle fingers away from this switch the FG makers put it on the back panel, not the front.

So if you have this switch in the "isolated" position, neither of the signal section's outputs (BNC shield and inner conductor) are connected back through the chassis and mains line cord to the mains ground.  Unless... you have connected the FG's output (or ext trigger, synch, whatever) BNC through a patch cable to some other instrument whose BNC connectors _are_ permanently actually grounded to that instrument's chassis which in turn is connected through its line cord to the mains ground and hence to all other stuff that is also connected to that mains ground. Like an oscilloscope probe "ground" reference clip.   :-BROKE
« Last Edit: January 09, 2020, 11:11:35 am by alsetalokin4017 »
The easiest person to fool is yourself. -- Richard Feynman
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf