PicoScope 7 Software
Available on Windows, Mac and Linux
Bursts of noise with 4444/441/PicoLog6
Bursts of noise with 4444/441/PicoLog6
I'm getting bursts of noise using 4444 scope, 441 probes, and PicoLog6 SW. The bursts occur approximately 18hrs apart (see pic). I'm certain it has nothing to do with the computer or the circuit. I saw a similar post in which the cable was at fault, but the scope and timing were different. Any ideas?
- Attachments
-
Re: Bursts of noise with 4444/441/PicoLog6
Hi Paul,
Before doing anything else, something similar caught me out for a little while, so, as a sanity check, is Channel B on Autorange for the input signal? If so, then check that it's not switching between ranges because the signal is going out-of-range (easily done when logging a noise signal or noisy signal because the PicoLog 6 display doesn't show the true peaks of the fast transients in a noise signal).
If its not the autoranger, then it would be worth knowing what you have channels A and B connected to (you could post some images of your setup), as they are clearly acting independently.
If there is a good reason why Channel A is not suffering from the same problem then it would be worth trying to esatblish where the interference on Channel B is originating from, for which you could do a little process of elimination. Try attaching an unconnected probe to see if you're getting RF pickup, or try running from a battery powered Laptop to see if was mains bourne.
Regards,
Gerry
Before doing anything else, something similar caught me out for a little while, so, as a sanity check, is Channel B on Autorange for the input signal? If so, then check that it's not switching between ranges because the signal is going out-of-range (easily done when logging a noise signal or noisy signal because the PicoLog 6 display doesn't show the true peaks of the fast transients in a noise signal).
If its not the autoranger, then it would be worth knowing what you have channels A and B connected to (you could post some images of your setup), as they are clearly acting independently.
If there is a good reason why Channel A is not suffering from the same problem then it would be worth trying to esatblish where the interference on Channel B is originating from, for which you could do a little process of elimination. Try attaching an unconnected probe to see if you're getting RF pickup, or try running from a battery powered Laptop to see if was mains bourne.
Regards,
Gerry
Gerry
Technical Specialist
Technical Specialist
Re: Bursts of noise with 4444/441/PicoLog6
Hi Gerry,
Both channels are on fixed ranges. The circuit is a buck converter with channel A connected across the load and channel B connected across the battery (each signal is actually 10x what's shown -- I got the multiplier wrong). The anomaly doesn't appear when using PicoScope6, but does with PicoLog6. So it's very confusing.
Thanks,
Paul
Both channels are on fixed ranges. The circuit is a buck converter with channel A connected across the load and channel B connected across the battery (each signal is actually 10x what's shown -- I got the multiplier wrong). The anomaly doesn't appear when using PicoScope6, but does with PicoLog6. So it's very confusing.
Thanks,
Paul
Re: Bursts of noise with 4444/441/PicoLog6
Hi Paul,
It's quite likely that you're not seeing this in PicoScope 6, because the anomaly occurs after a long time period (approximately19 hours according to your PicoLog 6 data plot) while PicoScope 6 can only capture data for an absolute maximum of just under 14 hours, when using the slowest sample rate, and for a significantly shorter time than that if you're using an appropriate sample rate for the output of a Buck converter.
As you're capturing the output of a Buck converter on channel B then I would assume the anomaly could be related to what's happening in the buck converter because, if you look closely, there are small bursts of noise on Channel A that coincide with the large bursts of noise on the battery. In PicoLog 6 you may not be sampling fast enough to capture the full peak-peak fluctuations at the converter output, so, it's possible that the fluctuations in converted power are related to (slower, and therefore more visible) fluctuations in supplied voltage.
A quick search on the internet revealed the following:
A converter attempting to operate with a conversion ratio of 1 – ρQ1 ≤ V2 ⁄ V1 < 1 will do so by oscillating between Buck and Boost modes, with a frequency that varies with the precise desired value of V2 / V1. This may excite any resonances in the system, which the controller will struggle to damp because of the collapse of control loop gain for 0 ≤ pn ≤ MρQ3. The result will be poor regulation for V2 / V1 in the indicated range. Similarly whenever the controller switches the converter from Buck to Boost modes, pn will be forced to slew from negative to positive. At the pn = 0 step discontinuity shown in Figure 5.8 system open-loop step responses will be triggered, which the controller will be unable to respond to due to its entry into the zero gain region 0 ≤ pn ≤ MρQ3. The result will be an apparent burst of noise or ringing in otherwise regulated quantities during such transitions.
This may or may not be the cause, but I would not rule out what is going on in the converter (also, the bursts are a bit too regular to be associated with a cable problem, but could also be a form of Electromagnetic Interference from some other device).
Regards,
Gerry
It's quite likely that you're not seeing this in PicoScope 6, because the anomaly occurs after a long time period (approximately19 hours according to your PicoLog 6 data plot) while PicoScope 6 can only capture data for an absolute maximum of just under 14 hours, when using the slowest sample rate, and for a significantly shorter time than that if you're using an appropriate sample rate for the output of a Buck converter.
As you're capturing the output of a Buck converter on channel B then I would assume the anomaly could be related to what's happening in the buck converter because, if you look closely, there are small bursts of noise on Channel A that coincide with the large bursts of noise on the battery. In PicoLog 6 you may not be sampling fast enough to capture the full peak-peak fluctuations at the converter output, so, it's possible that the fluctuations in converted power are related to (slower, and therefore more visible) fluctuations in supplied voltage.
A quick search on the internet revealed the following:
A converter attempting to operate with a conversion ratio of 1 – ρQ1 ≤ V2 ⁄ V1 < 1 will do so by oscillating between Buck and Boost modes, with a frequency that varies with the precise desired value of V2 / V1. This may excite any resonances in the system, which the controller will struggle to damp because of the collapse of control loop gain for 0 ≤ pn ≤ MρQ3. The result will be poor regulation for V2 / V1 in the indicated range. Similarly whenever the controller switches the converter from Buck to Boost modes, pn will be forced to slew from negative to positive. At the pn = 0 step discontinuity shown in Figure 5.8 system open-loop step responses will be triggered, which the controller will be unable to respond to due to its entry into the zero gain region 0 ≤ pn ≤ MρQ3. The result will be an apparent burst of noise or ringing in otherwise regulated quantities during such transitions.
This may or may not be the cause, but I would not rule out what is going on in the converter (also, the bursts are a bit too regular to be associated with a cable problem, but could also be a form of Electromagnetic Interference from some other device).
Regards,
Gerry
Gerry
Technical Specialist
Technical Specialist
Re: Bursts of noise with 4444/441/PicoLog6
Hi Gerry,
Thanks for the thoughtful and detailed response. I will look definitely look into it more deeply. I cannot interrupt the test right now (battery drain). Is it possible to run PicoLog6 and PicoScope6 simultaneously with one program using two channels and the other program using the other two channels? That would be useful in general, since it would free up my scope for other work while doing long logs like this.
Thanks,
Paul
Thanks for the thoughtful and detailed response. I will look definitely look into it more deeply. I cannot interrupt the test right now (battery drain). Is it possible to run PicoLog6 and PicoScope6 simultaneously with one program using two channels and the other program using the other two channels? That would be useful in general, since it would free up my scope for other work while doing long logs like this.
Thanks,
Paul
Re: Bursts of noise with 4444/441/PicoLog6
Hi Paul,
Unfortunately, you can only use one form of PicoTech software communication, to a Picotech hardware Data Acquisition device. So, you can only use PicoScope 6, or PicoLog 6 or your own software application (built with the SDK) to communicate with a PicoScope or PicoTech Data Logger. This is done deliberately to avoid conflicts in accessing the hardware.
Regards,
Gerry
Unfortunately, you can only use one form of PicoTech software communication, to a Picotech hardware Data Acquisition device. So, you can only use PicoScope 6, or PicoLog 6 or your own software application (built with the SDK) to communicate with a PicoScope or PicoTech Data Logger. This is done deliberately to avoid conflicts in accessing the hardware.
Regards,
Gerry
Gerry
Technical Specialist
Technical Specialist