I am currently trying to get my Picoscope 2205A up and running. Using the windows version (6.12.9.2917) at the moment but seem to have the same problems with the Linux version.
The circuit I am performing measurements on is a bench top power supply at 7.7 volts DC as input to a LM317 voltage regulator with an output of 5.11 volts. Both these values are stable and confirmed with a multi-meter. I have connected channel A to the voltage regulator output and channel B directly to the power supply. Both probes are set to 1x and the channels are set to 10volts/div.
The issue seems to occur when measuring both channels at the same time. The screen capture of the two waveforms can be seen below:
During this screen capture in the beginning I only have the channel A probe attached to the circuit. Next I attached the channel B probe. Finally I remove the channel A probe from the circuit. The following values were measured in each of these stages:
Can someone explain to me why this might be happening, or perhaps what I am doing wrong? When only one probe is attached and measuring it appears to give me exactly the expected values. If only I could get these values with both probes measuring.
What is even more strange, is that if I am measuring both channels as described above, and I select the range for Channel A to be +/-5 volts and leave Channel B at +/- 10 volts, the voltage offset to the real value gets even larger. At this point the measurement is:
In the original test I had both of the ground alligator clips clipped together on the ground output of the bench top power supply.
With the same conditions I attempted to test the circuit with only one ground clip attached (Channel A). In the screenshot below the beginning of trace I only had channel B connected to the power supply, with only the ground from channel A attached. Shortly afterwards I attached the channel A probe. You can see the same voltage offsets occur.
Is this an issue with the oscilloscope, or are my expectations unreasonable?
What is the load on your regulator? As a sanity check could you do the following:
[a] Connect the multi-meter to measure current between the PSU and the regulator (i.e. connect the regulator indirectly through the multi-meter instead of directly) then connect the probe tip and ground for channel A to the output and ground of the regulator, start a capture and save the psdata file. Disconnect the multi-meter from the PSU, connect the probe tip and ground for Channel B to the PSU and start a capture. Then, as in (a), with the probe for channel A connected to the regulator output, and the multi-meter in-line on the input (to measure current) connect the multi-meter and save another psdata file.
[c] Post the data files (single level voltage measurement for [a], and dual level voltage measurement for ) along with both current measurements for the regulator.
The behaviour of a regulator is going to depend upon how you're using it, loading it and driving it, so could you also post a diagram of your circuit components (with values) and connections.
Thanks for the help. I have conducted the tests as you described and have attached the two captured waveforms. Unfortunately this also shows the same behavior as described before. The circuit I am running at the moment is just a voltage regulator (LM317) setup to provide a little more than 5volts to feed an LM386 amplifier. Amplifier is working, sound comes out of the speaker. The circuit schematic is pictured below:
There is probably more than one mistake on the schematic, but the obvious one I just noticed is the R4 resistor is not a 680 kOhm resistor, it is just 680 Ohms. Otherwise that wouldn't work with the 5 volt output.
My previous request for your setup diagram, and sanity tests were just to make sure that the regulation was set up correctly (which it is in isolation, with a DC supply).
Now, if you look at the internals of the LM317 (page 8 & 10 of the T.I. data sheet here: http://www.ti.com/lit/ds/symlink/lm317.pdf) you can see that the input of the regulator's amp has across it a Zener voltage reference, in series with with a constant (but small) current source, and your resistor R4. When you probe the input of the regulator with a x1 Probe, you're effectively connecting a 1MΩ resistor in parallel with these components, providing an alternative path for current flow.
So, it's clear that what you are seeing is just the effect of a reduction in the current flowing through a Zener voltage reference (remember that the current of the current source is already very small, so the reduction is significant enough). The current is reduced by the amount that flows through the probe and Scope input channel, when you connect the probe. If you look at the diagram of reverse current flow for a Zener (http://cloud1.opensystemsmedia.com/5584 ... +Chris.jpg), you will see that as the current reduces, so does the voltage, and this is verified by the trace that you captured.
Under normal circumstances you would not be connecting anything across the input of the regulator, so you don't have to worry about any reduction in headroom for operation with a normal load.
Wow, that is a great explanation. I didn't think about the affect the probe was having on the input to the voltage regulator. I started measuring other voltage levels downstream from the LM317 while measuring the output of the voltage regulator and it stayed at a constant voltage as expected.
Thanks again for all the help. I am excited to keep experimenting.