## Info Please on Operation of AWG Channel on Picoscope 2405A

Post any questions you may have about our current range of oscilloscopes
musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Info Please on Operation of AWG Channel on Picoscope 2405A

First, I am a mechanical engineer so please bear with me as the explanation I am looking for is probably obvious to junior electronic engineers!

I setup a passive low-pass filter arrangement on a breadboard using a single resistor (160ohm) and capacitor (1micro-farad) to give a cut-off frequency just below 1kHz.

I connect a probe from the output of the AWG channel, connecting the positive clip to the "input leg" of the resistor and the negative croc clip to the "outer leg" of the capacitor - this provides the input voltage to the low-pass circuit. I have a BNC adaptor (single to double BNC connections) connected at the output of the AWG channel so that I can take a separate signal to channel A of the picoscope so I can see what signal the AWG channel is producing. I then connect the other probe to the channel B BNC connection - the other end of this probe is connected across the capacitor to give the output voltage of the low-pass circuit.
I set both probes to 1X - I understand that circuits can be loaded and expected to have to set the probes to 10X but do not get any sensible results when I do this.
Anyway, with the above settings I set the AWG voltage at +-1Volt and find that at low frequencies (30Hz) the channel A and B voltages are as expected ie they are both around +- 1 Volt.
Now, the crunch, as I increase the frequency I do find that the channel B voltage decreases significantly as I approach 1kHz - However, the channel A voltage is also decreasing!!!! It stabilises at a much lower value at higher frequencies.

I have attached a pdf file which shows the waveforms A & B at 3 different frequencies.

My question is why is the channel A (ie input to the low-pass circuit) voltage decreasing with increasing frequency? I expected this to remain constant irrespective of frequency?

Any help would be appreciated.
Attachments

Martyn
Posts: 3655
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

I would be connecting the output of the AWG directly to the circuit using a BNC adapter or similar, not using a probe.
Martyn
Technical Support Manager

musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Thanks Martyn,

I am a little confused. Could you please expand on your advice? Are you saying the probes supplied with the picoscope are the problem - and if so, why?

By recommending using a BNC adaptor do you mean using cables where one end has a BNC connector and the other end has crocodile clips? What would be the difference between this and the probes supplied with the picoscope?

It is not easy for me to fit BNC connections to my low-pass circuit as I am using a simple breadboard. However, if I can understand why your suggestion would work then I would make all attempts to do this - I just need some explanation of what is happening with my present setup and why your recommendation would work.

Martyn
Posts: 3655
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

That would be the wrong way round, you need to use the BNC end connected to the AWG output of the scope, and the croc clips onto your circuit, so you have a piece of wire connecting the two together.
Martyn
Technical Support Manager

terrypin
Newbie
Posts: 0
Joined: Mon Jun 16, 2014 4:23 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

So frustrating to 'Preview' a carefully prepared post and get its contents deleted instead!

I tried again, this time previewing directly after that sentence above plus an image. Same result, although now I got a message: "Error 504 Ray ID: 461d1937e8516a97 • 2018-09-29 08:33:19 UTC
Gateway time-out"
and a diagram showing all was OK with my Browser, and at 'Cloudfare' in London, but a 'Host error' at www.picotech.com

I'll try a third time, this time posting a link instead of embedding the image.

I wasn't clear what you meant by 'input leg' and 'outer leg', but given that odd result I think you must have connected wrongly. I'm not in my shed and my 2205A disappointingly has no AWG, but this shows how I'd connect.

https://www.dropbox.com/s/7nl3x4f33g4x9 ... .jpg?raw=1

IOW, just make sure your signal (and therefore Ch A) goes across both the R and C, and your Ch B is taken from across C.

IMO it's irrelevant whether you use probes or croc-clips, at least at these frequencies. In fact with small or tightly spaced parts, probes are sometimes the only practical option. And those two-way BNC adapters are very handy for exactly the purpose you have there.

P.S. To web manager: With the security of a copy pasted elsewhere, I tried my embedded image once more, with same error message. This site plainly has a problem with images.

musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Martyn,
Obviously my explanation of how I connected from the picoscope to my R-C circuit wasn't good enough. I did use the crocodile clips to connect to the circuit and BNC connector to the WAG output - I can't see how I could, from a practical point of view, connect it the other way around?

I have connected exactly as terrypin has drawn. I have now added a photo which hopefully will confirm this. I also agree with terrypin about the difference between crocodile clips and BNC connections at the circuit for such low frequencies would be expected to be neglible, assuming I get decent connections at the crocodile clips. That's why I asked for an expansion of your recommendation on why I should do what you suggested? I am a humble mechanical engineer and very willing to learn.

terrypin - could you please explain what devices and connections you used to get the waveforms that you showed? I really don't understand why my AWG output signal (also taken to channel A) can change (ie decrease) so much with increasing frequency - could it be something to do with me taking the same signal from AWG output to channel A?

musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

It looks like the attachment of my .jpg file did not work so I have put the photo into a .pdf file and tried to attach that here.

I look forward to any responses to my queries that anyone may have.
Attachments
R_C Circuit Connections.pdf

terrypin
Newbie
Posts: 0
Joined: Mon Jun 16, 2014 4:23 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

I couldn't get to my shed workshop at the time so my screenshot was a simplified simulation intended to show you the correct cable connections. And anyway, as mentioned, my 2205A apparently doesn't have an operational AWG signal socket. So when I get to the shed later I'll breadboard your RC filter with an independent signal generator.

The reason for your input signal decreasing in amplitude (although not at the same rate as your Ch B filtered signal) will be because the AWG will have some output impedance. That will form a voltage divider with the combined impedance of R and C. My simplistic simulated signal generator, in contrast, had no output impedance.

When I breadboard it shortly I expect my own signal generator will also show the same sort of result as you had. I'll report back with details.

Bottom line IMO: your test setup is fine.

terrypin
Newbie
Posts: 0
Joined: Mon Jun 16, 2014 4:23 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Attached are screenshots from bread-boarding your circuit at three frequencies: 70Hz, 1kHz and 10kHz. Lacking a 160 ohm I used a 150 ohm resistor. And as per my post early this morning I used an external signal generator.

As expected the results are similar to yours. I haven't done the math but they look broadly consistent with the specified 50 ohm output impedance of my signal generator. If your design allows, increase R from 160 to 1600 and reduce C from 1.0 to 0.1 uF. That will maintain Ch A very close to 1Vpp.

P.S. to Pico Support: I used 'Attachments' rather than the 'Insert image' tool. The failure of the latter which I reported two days ago remains unchanged today. Is it supposed to work?
Also, there appears to be a size limit for an attachment - but what is it?

Attachments

musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Hi Terry,

Thanks very much for your excellent explanation of what is likely to be happening. Just to let you know (as this may mean a lot more to you than me) the picoscope manual states that the AWG output impedance is 600ohms while the first three input channels (A,B & C) have an impedance of 1Megaohm.

I am still confused about some things though that, if you don't mind, you might be able to answer?

1. As it stands it looks like using my picoscope with an R-C circuit like the one I have setup there is no way to maintain a constant input signal of 1V across a frequency range of 0-1khz. Is there some circuit I can put between the AWG output and my circuit that would ensure this constant voltage input?

2. Would I be right in thinking that if I can get the ratio of R-C circuit output to R-C circuit input (ie ChB/ChA) then I could at least get the true attenuation of the low-pass circuit?

3. My real concern is that I was planning to use this circuit on another project I am working on and my testing of the circuit using the picoscope was purely to see that the R-C circuit I created would do the job I am looking for. To let you understand, I have a setup that uses a variable speed drive. The drive produces electromagnetic interference due to pulse width modulation in the drive. This interference radiates to measurement instrumentation I am using (a 1 per rev signal from a rotating shaft and accelerometers to measure vibration). I was hoping the low-pass filter with the R-C parameters I have selected would reduce the high frequency noise that is corrupting my measured signals, to an acceptable level. However, I am now wondering if the measurement signals I feed into the R-C filter will be attenuated as the AWG one is? If this happens I would have no idea how the calibration figures for my sensors would be affected. Is there something I can do to avoid this happening in such a situation?

Sorry for all the questions but I didn't expect this one to be so tricky and I can see you seem to be well versed on the subject matter.

Thanks again.

Irvin

musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Terry,

Just as I sent my last post around 13:55 I noticed your more recent post (12:38pm) - so thank you very much for that. I think I understand it a little better now. If you don't mind could you still have a look at my queries and if you can assist in any way that would be great.

Thanks

Irvin

terrypin
Newbie
Posts: 0
Joined: Mon Jun 16, 2014 4:23 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Hi Irvin,

I’ve only now (08:15 Tuesday) seen your reply to my two posts yesterday morning and lunch time. Your main reply is time-stamped 13:55 Monday, yet wasn’t present when I checked throughout the day, up until 22:30. Presumably because of moderator delay. So who knows when you’ll see this one!

I have to go out very shortly so I’ll have to be brief. I can follow up later this evening if necessary.

I’m just a hobbyist, so hopefully others might step in and offer advice based on professional know-how.

The key point underlying your Picoscope problem is the one I made about the voltage divider. So lowering the output impedance of your test signal will make a big difference. Zero is impossible but an amplifier could get it to a few ohms. Think of the 4 or 8 ohm load of speakers.

Also, the filter itself should have as large an input impedance as possible. Did you try the change I suggested?

More important, I suspect that the simple RC passive filter might not deliver the broader objective you seek. But (if we haven’t done so already!), we are now straying well OT for this forum. Are you also using any of the electronics forums or newsgroups?

Hopefully on my return some six or eight hours later today I’ll see your reply. We can then either continue by email or I’ll suggest other resources.

musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Hi Terry,

Thank you very much for all of the information you have provided. I agree that I have probably strayed way beyond the subject matter of this forum.

I don't know which other forums I should approach but will look into this.

I will definitely try the R-C values you suggested and am grateful for that suggestion.

Your comments about the filter input impedance and signal-in output impedance are very important and I think this is where my understanding is severely lacking and I need to learn a lot more. Perhaps you know of some good references (online or otherwise).

I also agree that my filter may not even be sufficient to achive my aims for my project but wanted to try until I met these other unforeseen (for me!) problems.

Thanks once again for your advice - it has certainly helped me understand the situation a little more.

Irvin

terrypin
Newbie
Posts: 0
Joined: Mon Jun 16, 2014 4:23 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Hi Irvin,

Thinking more about your requirement reinforces my view that using a passive RC low-pass filter may prove unsatisfactory. But without knowing much more about the project and particularly the unwanted interference I can't really be of much help. Was the PWM circuit your own design? If so it sounds as if you already do have a fair grasp of electronics!

If it's largely filter design that you're struggling with, then googling "how to design low pass filters" and selectively browsing some of the 400,000 odd 'hits' will soon get you up to speed. Accompanied by actual experimenting, of course, ideally with the real interference signals, not a sine wave from your AWG. Or narrow it down with a more specific google search, bringing in keywords like "PWM" or "PWM interference".

BTW, re the AWG, I have you to thank for something. I was sure I'd read a Pico Technology web page comparing product features and concluded that my 2205A had no AWG. But today, prompted by our discussion, I tried it nevertheless - and was very pleased to discover that it does.

I believe some of the Picoscopes have built-in filter functionality and wonder if your 2405A, a more expensive model than mine, is one of them?

As for other electronics forums and newsgroups, there are many. So on that too I'd suggest you do some googling and then lurk in a few until you find one appropriate to your experience and interests.

It might also help you to post a more detailed description of your project in the Projects section of this forum, raising specific questions about using the Picoscope and its software.

Best wishes,

musicman
Newbie
Posts: 0
Joined: Thu Sep 01, 2016 8:10 pm

### Re: Info Please on Operation of AWG Channel on Picoscope 2405A

Hi Terry,

Great recommendations from you - I will certainly follow up on many of your suggestions.
I greatly appreciate the assistance you gave me and was pleased to see that your picoscope has an AWG!

The PWM problem I am experiencing is coming from a variable frequency controller and AC induction motor arrangement I purchased. The PWM effect apparently comes from the need to change the power input from AC to DC and back again so that variable frequency can be employed to produce a variable speed motor! Apparently the electromagnetic interference is an unwanted "side-effect" of such a drive.

I had checked previously to see if my picoscope included filtering but unfortunately this feature is only available in much more pricey models.

Believe it or not I was also coming to the conclusion that I might need a more efficient low-pass filter perhaps using a couple if stages and op-amp for increased gain - but that is another project for me!

Many thanks again and Best Wishes to you.

Irvin