It’s a shame I can only give one thumb up! EXCELLENT video that explains just what I was looking for. Thank you so much for dedicating your time to explaining this.
You're welcome - and thank you for the praise! Regarding the dedication of my time: It was actually a lot of fun - it's always nice to produce interesting pictures on the scope and digging out my old (still working) amplifier was a blast ;-)
It's also available as OWON XDS3204AE. I don't know who's the OEM. I've three videos about that thing: Unboxing: ua-cam.com/video/pYrK1Ge2Tr0/v-deo.html Review 1/2: ua-cam.com/video/lYL1LuMJ6Uo/v-deo.html Review 2/2: ua-cam.com/video/EqCxT5vQNqU/v-deo.html A fair warning: These are pretty long and winded (I go through each and every function of that scope).
You are truly the... "Frequency Whisperer". A brain bending foray for a neophyte like myself. However, I will keep reaching for the moon. Thanks so much.
I thank you for the praise! But I really don't consider myself a "frequency whisperer" 😉 I just sat down and played with the thing until I could show something pretty on the scope. And please, do keep on reaching for the moon!
I have found another way to achieve this that I believe is more accurate. Simply pass the trigger output through a low-pass filter and a peak detector. This will produce a signal that has a rising edge only at the start of the sweep. This you can use to trigger the scope. To produce a log log diagram you can set the sweep to be logarithmic and you can use the math channel of the scope to log the y-axis. I did this very simply with passive components on a breadboard. I.e RC filter + peak detector using diode capacitor and resistor. It also helps to set the "hold-off" time for the scope trigger to something close to the sweep frequency.
@@robertssmorgasbord Ha, true. It is cheating. But none the less useful. Because I think most viewers have a couple of capacitors, diodes and resistors laying around.
@@davidosterberg But then, it's not cheating ... if it yields better results :-) Any yes, I guess everybody can cobble together a RC low pass, a diode peak detector and two BNC sockets if the need arises - no PCB necessary. Put the whole thing into a copper or steel tube with the BNC sockets sealing the openings and you're done.
You're welcome! You can for sure make some qualitative frequency response measurements with the Poor Man's "Bode Plot". At the very least it will be fun!
Hi Robert! Everything Sunday morning your video is a treat. Thank you so much for such a comprehensive explantation an problem solving. . Since I have a FY6900 model I will check to see if the offset problem has been corrected or not on that earlier model. My opamp amp has a wide band to cover (DC to 1 MHz). Now that I have that great septup that I can use to caracterize my amplifier, I can test its frequency response. Your great result opens another question/problem to solve: a peak detector. Is there such thing that would cover the above -mentioned band that could be done? ... A next video may be ? Thank you soooo much for all. Yours, RJM.
Hi René-Jean, thanks for the praise! Regarding a peak detector - EEVblog has already a nice peak detector tutorial: ua-cam.com/video/jllsqRWhjGM/v-deo.html To be honest, I don't know when I'll come around doing something along those lines. But I've put a related Post-It on my "wall of ideas" - no promises though. Best Regards Robert
I'm using the unit for characterize IF transformers on old vacuum tube receivers....frequency is from 450 to 500 khz and the solution (for me) was to use the vco function with ch2 saw tooth as sync gen. And it works pretty good....
I guess a frequency sweep of ±5.3% (25 kHz / 475 kHz) is much easier on the unit than trying to sweep the whole audio range (±99.8%, 9.99kHz / 10.01kHz) like I did :-) Good thing it works for you in that application.
The frequency response is pretty easy, but the phase response without measuring the waveform is a bit harder to automate. I think the way I would go about this, is to dump those values you had into excel, and get that to plot the bode. So if we measure time between peaks for each period viola! (now try programming that in excel!) Maybe you could do a quick freq response on some scopes by sweeping the frequency on FFT with persistence on.
Well, you're absolutely right. However, the challenge here was to create some kind of frequency response plot (poor man's "Bode plot") with just that cheap signal generator and a very simple oscilloscope (no FFT or other advanced features). The whole thing was actually a requested by a viewer (comment on one of my FY6800 review videos). Regarding FFT with persistence on: That's definitely something to try.
@@robertssmorgasbord Well yeah, my comment was not a critic, I think all these videos help those starting their electronics journey and sometimes gives ideas to the pros. I am just refreshing my skills, so going to try measurement from passive filters and build up against their laplace equivalents for frequency and time domain responses. Thanks for the video.
@@suzesiviter6083 No sweat! I didn't feel criticized and your suggestions are - as I already said - absolutely valid. It's just that the video has another scope and I wanted to point that out 😉
@@robertssmorgasbord No worries, just an update, been doing frequency responses today, with my Siglent SDS1202x-e I found that the best way to plot frequency response is using FFT with peak-hold, then adjusting the sig gen sweep time such that the FFT has time to record samples without skipping, next I will try the phase using X-Y function and some maths.
@@suzesiviter6083 Sounds interesting! Unfortunately the FFT in my BNC P4025T (really an OWON XDS3204AE) is very basic ( ua-cam.com/video/EqCxT5vQNqU/v-deo.html ). The FFT of your SIGLENT SDS1202X-E does look sophisticated compared to that. Anyway, I'm very interested in your results you get and the actual settings you're using.
Hi Robert, this was a fun video to watch! I appreciate you got this box from the past to have some fun with. I have a FY6200 which is the panel version, does not have the coax connectors on the back, just power and USB, although in the manual which is FY6200 specific they talk about the VCO etc. I asked FeelTech in the chat and they replied, quote "There is no definition on the back. The customer can connect the corresponding signal as required. VCO signal input from the rightmost terminal of the panel". What do you think? Could they have the counter input jack configured in such a way that when using the VCO function that input cold be used as such? My scope is a 2 channel Hantek DSO5102P so I would have to drop the input of the DUT.
Well, sounds like they're really talking about using the counter input (rightmost BNC) as VCO input. And yes, that's absolutely feasible: That BNC would just have to feed into the counter analog front end as well as VCO analog front end. Just test it. You can't possibly destroy the counter input by applying 0V to 5V DC to it (limits of the VCO input). Configure channel 1 as VCO, apply some positive voltage (e.g. 1.5V from a battery) to the rightmost BNC and see if the output frequency of channel 1 changes. Not monitoring the input of the DUT is not a big deal. You could connect your scope temporarily to the input of the DUT just to set the vertical scale of the scope.
I was just working out to create the same Bode Plot with an identical FY6800. Great work!
You're welcome :-)
It’s a shame I can only give one thumb up! EXCELLENT video that explains just what I was looking for. Thank you so much for dedicating your time to explaining this.
You're welcome - and thank you for the praise! Regarding the dedication of my time: It was actually a lot of fun - it's always nice to produce interesting pictures on the scope and digging out my old (still working) amplifier was a blast ;-)
Thank you, i'm doing same things based on your calculation. Now i'm able to graph curves similar as yours on my cheap hantek dso2d10+fy6900. 😊
You're welcome 😃 Function generators can be so much fun 😉
"BNC" first time ever I have seen this make of scope. I only had seen Tektronix, Agilent, HP, Lecroy, R&H etc.
It's also available as OWON XDS3204AE. I don't know who's the OEM. I've three videos about that thing:
Unboxing: ua-cam.com/video/pYrK1Ge2Tr0/v-deo.html
Review 1/2: ua-cam.com/video/lYL1LuMJ6Uo/v-deo.html
Review 2/2: ua-cam.com/video/EqCxT5vQNqU/v-deo.html
A fair warning: These are pretty long and winded (I go through each and every function of that scope).
You are truly the... "Frequency Whisperer". A brain bending foray for a neophyte like myself. However, I will keep reaching for the moon. Thanks so much.
I thank you for the praise! But I really don't consider myself a "frequency whisperer" 😉 I just sat down and played with the thing until I could show something pretty on the scope. And please, do keep on reaching for the moon!
I have found another way to achieve this that I believe is more accurate. Simply pass the trigger output through a low-pass filter and a peak detector. This will produce a signal that has a rising edge only at the start of the sweep. This you can use to trigger the scope. To produce a log log diagram you can set the sweep to be logarithmic and you can use the math channel of the scope to log the y-axis.
I did this very simply with passive components on a breadboard. I.e RC filter + peak detector using diode capacitor and resistor. It also helps to set the "hold-off" time for the scope trigger to something close to the sweep frequency.
A very simple and elegant solution! But it's kind of cheating using extra hardware :-)
@@robertssmorgasbord Ha, true. It is cheating. But none the less useful. Because I think most viewers have a couple of capacitors, diodes and resistors laying around.
@@davidosterberg But then, it's not cheating ... if it yields better results :-) Any yes, I guess everybody can cobble together a RC low pass, a diode peak detector and two BNC sockets if the need arises - no PCB necessary. Put the whole thing into a copper or steel tube with the BNC sockets sealing the openings and you're done.
Very interesting. Can you describe the form of the sync output? eg it's waveform and timing.
thank a lot, I'm wondering about how to set up my benchmark table for do some freq. respond check. this vdo is help me a lot.
You're welcome! You can for sure make some qualitative frequency response measurements with the Poor Man's "Bode Plot". At the very least it will be fun!
Hi Robert! Everything Sunday morning your video is a treat. Thank you so much for such a comprehensive explantation an problem solving. . Since I have a FY6900 model I will check to see if the offset problem has been corrected or not on that earlier model. My opamp amp has a wide band to cover (DC to 1 MHz). Now that I have that great septup that I can use to caracterize my amplifier, I can test its frequency response.
Your great result opens another question/problem to solve: a peak detector. Is there such thing that would cover the above -mentioned band that could be done? ... A next video may be ? Thank you soooo much for all. Yours, RJM.
Hi René-Jean, thanks for the praise!
Regarding a peak detector - EEVblog has already a nice peak detector tutorial: ua-cam.com/video/jllsqRWhjGM/v-deo.html
To be honest, I don't know when I'll come around doing something along those lines. But I've put a related Post-It on my "wall of ideas" - no promises though.
Best Regards
Robert
I'm using the unit for characterize IF transformers on old vacuum tube receivers....frequency is from 450 to 500 khz and the solution (for me) was to use the vco function with ch2 saw tooth as sync gen. And it works pretty good....
I guess a frequency sweep of ±5.3% (25 kHz / 475 kHz) is much easier on the unit than trying to sweep the whole audio range (±99.8%, 9.99kHz / 10.01kHz) like I did :-) Good thing it works for you in that application.
@@robertssmorgasbord you are right....but for price we have paied for is good enough
@@SalvoMorina And you are absolutely right too! The value offered to a hobbyist for that price is indeed good.
Thank you for this information. Have you tried this at say 10.7 mHz?
@@jonka1 no, but I think that it works
The frequency response is pretty easy, but the phase response without measuring the waveform is a bit harder to automate.
I think the way I would go about this, is to dump those values you had into excel, and get that to plot the bode.
So if we measure time between peaks for each period viola! (now try programming that in excel!)
Maybe you could do a quick freq response on some scopes by sweeping the frequency on FFT with persistence on.
Well, you're absolutely right. However, the challenge here was to create some kind of frequency response plot (poor man's "Bode plot") with just that cheap signal generator and a very simple oscilloscope (no FFT or other advanced features). The whole thing was actually a requested by a viewer (comment on one of my FY6800 review videos).
Regarding FFT with persistence on: That's definitely something to try.
@@robertssmorgasbord Well yeah, my comment was not a critic, I think all these videos help those starting their electronics journey and sometimes gives ideas to the pros.
I am just refreshing my skills, so going to try measurement from passive filters and build up against their laplace equivalents for frequency and time domain responses.
Thanks for the video.
@@suzesiviter6083 No sweat! I didn't feel criticized and your suggestions are - as I already said - absolutely valid. It's just that the video has another scope and I wanted to point that out 😉
@@robertssmorgasbord No worries, just an update, been doing frequency responses today, with my Siglent SDS1202x-e I found that the best way to plot frequency response is using FFT with peak-hold, then adjusting the sig gen sweep time such that the FFT has time to record samples without skipping, next I will try the phase using X-Y function and some maths.
@@suzesiviter6083 Sounds interesting! Unfortunately the FFT in my BNC P4025T (really an OWON XDS3204AE) is very basic ( ua-cam.com/video/EqCxT5vQNqU/v-deo.html ). The FFT of your SIGLENT SDS1202X-E does look sophisticated compared to that. Anyway, I'm very interested in your results you get and the actual settings you're using.
Pity the internal sweep does not generate and output sync pulse. :(
Indeed! But you get what you pay for ... high end signal generators have of course a sync output and input.
Hi Robert, this was a fun video to watch! I appreciate you got this box from the past to have some fun with. I have a FY6200 which is the panel version, does not have the coax connectors on the back, just power and USB, although in the manual which is FY6200 specific they talk about the VCO etc. I asked FeelTech in the chat and they replied, quote "There is no definition on the back. The customer can connect the corresponding signal as required. VCO signal input from the rightmost terminal of the panel". What do you think? Could they have the counter input jack configured in such a way that when using the VCO function that input cold be used as such? My scope is a 2 channel Hantek DSO5102P so I would have to drop the input of the DUT.
Well, sounds like they're really talking about using the counter input (rightmost BNC) as VCO input. And yes, that's absolutely feasible: That BNC would just have to feed into the counter analog front end as well as VCO analog front end. Just test it. You can't possibly destroy the counter input by applying 0V to 5V DC to it (limits of the VCO input). Configure channel 1 as VCO, apply some positive voltage (e.g. 1.5V from a battery) to the rightmost BNC and see if the output frequency of channel 1 changes.
Not monitoring the input of the DUT is not a big deal. You could connect your scope temporarily to the input of the DUT just to set the vertical scale of the scope.
Robert's Smorgasbord Thanks for your feedback Robert! I will test it.
@@MarcelHuguenin You're welcome! Let me know if it works. Other people might be interested too.