Diode snubbers that is -- I was able to set the conditions so that I could take a few snapshots of an somewhat optimally snubbed out diode ringing -- the power supply has a lot more inductance than will be found in the typical torroid supply -- I use a choke from the center tap to limit transients --
using too large a cap wastes energy -- perhaps more important when there is lower inductance and a higher ringing frequency
note, the time base in the lower snapshot is 400uSec, not 200.
using too large a cap wastes energy -- perhaps more important when there is lower inductance and a higher ringing frequency
note, the time base in the lower snapshot is 400uSec, not 200.
An externally hosted image should be here but it was not working when we last tested it.
AndrewT said:
What's your problem ??? I just happen to have a 'scope with a printer attached -- a TDS3012B -- the images are reduced to save bandwidth. The more faint image is an FFt. ..and FYI I did an entire summer as a lab-rat for a physicist before I learned you could make more money in other disciplines!
I noted that the lowest graph used the 400uSec timebase, since I wasn't going to solder the 1uF caps back across the diodes.
you can get an even clearer picture if you look at the diode ringing on a switch-mode supply since the turn-off is immediate (but the diode takes ever so slightly longer) -- this ringing just rides up and down the ripple.
Jack,
I think that Andrew was *not* disparaging your pictures. Instead I believe that he was commenting on how much ringing is present on an un-snubbered supply, and his amazement that we spend our time listening to equipment with that much noise.
I have a Tek 'scope from the same family as yours (also with the FFT module) and would like duplicate your measurements. Could you give a few more details on how the measurement was made? Mostly I'd like to know what probe you used, where you connected the ground, and where you looked at the signals.
I would assume that the bottom trace with the 1 uF caps used the same 6.2 k series resistor as the middle trace. Is this correct?
Thanks,
Charles Hansen
I think that Andrew was *not* disparaging your pictures. Instead I believe that he was commenting on how much ringing is present on an un-snubbered supply, and his amazement that we spend our time listening to equipment with that much noise.
I have a Tek 'scope from the same family as yours (also with the FFT module) and would like duplicate your measurements. Could you give a few more details on how the measurement was made? Mostly I'd like to know what probe you used, where you connected the ground, and where you looked at the signals.
I would assume that the bottom trace with the 1 uF caps used the same 6.2 k series resistor as the middle trace. Is this correct?
Thanks,
Charles Hansen
If you look at the time base, it is visible that the pulse with on the scope is ~ 1,6 mSec. The pulse height is ~540 mV.
This all looks like the setup what I have applied, and looks like if it were a charge current pulse, that we are looking at. The typical value seen by me was ~ 2- 3 msec.
This can be measured by putting a small value, say .1 ohm, .05 ohm reistor in series with one of the diodes, and measuring with two probes, in differential mode. You get small voltage across that resistor, what could be similar to the values seen here. [for example, in case of .1 ohm we see here 5 amps]
I have seen this same "snitch" also, this is where the diodes switch off.
If one were looking at the tension at the rectifiers, then one could see the 50 Hz sinus [pardon, 60 Hz at your place], with the peak of it "snipped off", like if it were an amp in clipping.
I hope i did not make a big mistake...
It is the authors task to explain clearly, I am only an intruder while he is missing..
This all looks like the setup what I have applied, and looks like if it were a charge current pulse, that we are looking at. The typical value seen by me was ~ 2- 3 msec.
This can be measured by putting a small value, say .1 ohm, .05 ohm reistor in series with one of the diodes, and measuring with two probes, in differential mode. You get small voltage across that resistor, what could be similar to the values seen here. [for example, in case of .1 ohm we see here 5 amps]
I have seen this same "snitch" also, this is where the diodes switch off.
If one were looking at the tension at the rectifiers, then one could see the 50 Hz sinus [pardon, 60 Hz at your place], with the peak of it "snipped off", like if it were an amp in clipping.
I hope i did not make a big mistake...
It is the authors task to explain clearly, I am only an intruder while he is missing..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.