Author Topic: Rigol dg1022u amplitude issue [solved]  (Read 4619 times)

0 Members and 1 Guest are viewing this topic.

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Rigol dg1022u amplitude issue [solved]
« on: December 22, 2017, 10:48:52 pm »
Hello..another topic about this function generator..is driving me to crazy..I have upgraded the firmware..but now I have amplitude issue..if I set 25Mhz, 10Vpp, HiZ and attach the probe to the function gen (to terminate it) I get the 25Mhz but 5 or 6 Vpp, have no idea why, what could be?
« Last Edit: December 23, 2017, 09:51:31 pm by NiaDebesis »
 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: Rigol dg1022u amplitude issue
« Reply #1 on: December 22, 2017, 11:02:38 pm »
Are you saying that if your reduce the frequency to 20Mhz, then you will get exactly the output voltage?

You should be using BNC to BNC connector.
Remember to set the oscilloscope coupling to 1:1
Oscilloscope should be set to 1megOhm input impedance example high impedance input. [Not 50 Ohms]

A picture of your setup would be helpful.
Explain your scope setting used.
« Last Edit: December 22, 2017, 11:15:59 pm by Armadillo »
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #2 on: December 23, 2017, 07:25:04 am »
Are you saying that if your reduce the frequency to 20Mhz, then you will get exactly the output voltage?

You should be using BNC to BNC connector.
Remember to set the oscilloscope coupling to 1:1
Oscilloscope should be set to 1megOhm input impedance example high impedance input. [Not 50 Ohms]

A picture of your setup would be helpful.
Explain your scope setting used.

I am pretty sure that I am doing something wrong..is to strange to be a hardware issue, today Ill try againg, I am using a 1054z, HighZ on the F.G and I use a probe to connect it to the oscilloscope, not a bnc to bnc, because I don't have any termination at the moment for the scope
 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: Rigol dg1022u amplitude issue
« Reply #3 on: December 23, 2017, 08:16:30 am »

I am pretty sure that I am doing something wrong..is to strange to be a hardware issue, today Ill try againg, I am using a 1054z, HighZ on the F.G and I use a probe to connect it to the oscilloscope, not a bnc to bnc, because I don't have any termination at the moment for the scope

Thanks for realizing that. Because the DG1022U cannot be considered as a cheapy, at around US$400. At this price, the output voltage need to be within tolerance specs, otherwise long time ago, it would had been condemned.

BTW, DG1022U included a BNC to BNC connector with the manual and disks in the box.
 :)
 

Offline Damianos

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: gr
Re: Rigol dg1022u amplitude issue
« Reply #4 on: December 23, 2017, 09:08:11 am »
Maybe the answer is somewhere in the manual.
Many (if not most) generators do not give their maximum amplitude to the maximum of their frequency range.
What is stated in the specifications? Does the device display a message when these settings are entered?
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #5 on: December 23, 2017, 09:16:35 am »
Maybe the answer is somewhere in the manual.
Many (if not most) generators do not give their maximum amplitude to the maximum of their frequency range.
What is stated in the specifications? Does the device display a message when these settings are entered?

The device output max 10Vpp at 25Mhz at 20Mhz and below 20Vpp
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #6 on: December 23, 2017, 09:20:27 am »

I am pretty sure that I am doing something wrong..is to strange to be a hardware issue, today Ill try againg, I am using a 1054z, HighZ on the F.G and I use a probe to connect it to the oscilloscope, not a bnc to bnc, because I don't have any termination at the moment for the scope

Thanks for realizing that. Because the DG1022U cannot be considered as a cheapy, at around US$400. At this price, the output voltage need to be within tolerance specs, otherwise long time ago, it would had been condemned.

BTW, DG1022U included a BNC to BNC connector with the manual and disks in the box.
 :)

I have a bnc to bnc but no termination for the scope, but using a probe, that have is own resistance..it would shown the right Vpp
But above 5Mhz it always shown a 2 time smaller voltage, I will buy a 50Ohm termination for the scope and try again, but, I really dom't know
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #7 on: December 23, 2017, 10:06:28 am »
The best way to solve this problem that I have found is to take a good input impedance. I have found a Tektronix Normalizer RC=1 Mohm x 33pF  for 70 bucks, not cheap, but haven't found nothing better
 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: Rigol dg1022u amplitude issue
« Reply #8 on: December 23, 2017, 10:18:26 am »
I should think this is more of a foundation problem than equipment problem.

Connect BNC output to Oscilloscope BNC input. As simple as that to test the DG1022U.
Either you set DG1022U to 50 ohms output and oscilloscope input to 50 ohms input
OR
Set both to high impedance mode.

I don't know what you talking about terminating the oscilloscope??

 ;)

 

Offline cs.dk

  • Supporter
  • ****
  • Posts: 642
  • Country: dk
Re: Rigol dg1022u amplitude issue
« Reply #9 on: December 23, 2017, 10:22:34 am »
I don't know what you talking about terminating the oscilloscope??

Rigol D1054z does not have an internal 50 ohms terminator/resistor.
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #10 on: December 23, 2017, 10:35:00 am »
I don't know what you talking about terminating the oscilloscope??

Rigol D1054z does not have an internal 50 ohms terminator/resistor.

Exactly
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #11 on: December 23, 2017, 10:45:18 am »
https://imageshack.us/i/pmFlftJuj = 5Mhz

https://imageshack.us/i/poY6uU9ij = 15Mhz

https://imageshack.us/i/pmMWL6Pvj = 25Mhz

Using a bnc to bnc, HighZ on F.G and no termination on the scope


 

Offline Damianos

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: gr
Re: Rigol dg1022u amplitude issue
« Reply #12 on: December 23, 2017, 12:02:03 pm »
It seems that is activated the bandwidth limit, on channel-1.
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #13 on: December 23, 2017, 12:34:42 pm »
It seems that is activated the bandwidth limit, on channel-1.

Deatctivated, tried on other channel, nothing, same problem
 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: Rigol dg1022u amplitude issue
« Reply #14 on: December 23, 2017, 12:48:49 pm »
I don't know what you talking about terminating the oscilloscope??

Rigol D1054z does not have an internal 50 ohms terminator/resistor.

Exactly

Can read the "Either" ....... "OR".......
Be more objective base to your own problems.  ;D
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #15 on: December 23, 2017, 12:58:57 pm »
I don't know what you talking about terminating the oscilloscope??

Rigol D1054z does not have an internal 50 ohms terminator/resistor.

Exactly

Can read the "Either" ....... "OR".......
Be more objective base to your own problems.  ;D

I am sorry, but the only solution for me is to have a HigZ on the scope, that now I don't have, If anyone have a better solution to solve this problem, better for me :)
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #16 on: December 23, 2017, 04:35:13 pm »
Nothing, I am out of ideas, the input Z of the scope is 1M, and the problem is still here, I have also done a reset of F.G (Dunno if the factory reset on tge rigol remove the hack or will cause problems)and still nothing
« Last Edit: December 23, 2017, 04:49:08 pm by NiaDebesis »
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #17 on: December 23, 2017, 07:12:20 pm »
?
 

Offline Damianos

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: gr
Re: Rigol dg1022u amplitude issue
« Reply #18 on: December 23, 2017, 07:37:30 pm »
Lets try a little more:
You use a coaxial cable without a proper termination. So, the dominant characteristic is its capacitance plus the inductance and the resistive losses. To verify if this is valid, set the generator to slow sweep from 10MHz to 25MHz and observe if there are "tops and valleys".
Try to find a terminator, for these frequencies even a simple resistor can do the job.
Also verify the compensation of your probe, assuming that you use a 10:1 type, an 1:1 is not proper in this case. Try to do this with a square signal around 2Vp-p and 1MHz. Use it directly at the output of the generator.
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #19 on: December 23, 2017, 07:41:17 pm »
Lets try a little more:
You use a coaxial cable without a proper termination. So, the dominant characteristic is its capacitance plus the inductance and the resistive losses. To verify if this is valid, set the generator to slow sweep from 10MHz to 25MHz and observe if there are "tops and valleys".
Try to find a terminator, for these frequencies even a simple resistor can do the job.
Also verify the compensation of your probe, assuming that you use a 10:1 type, an 1:1 is not proper in this case. Try to do this with a square signal around 2Vp-p and 1MHz. Use it directly at the output of the generator.

Ill try later! Thanls for the advice!
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #20 on: December 23, 2017, 07:43:15 pm »
Also..the Rigol have an HigZ input, of 1Mohm..soo..techincally is well terminated, and I am using athe probes that cames with the scope
 

Offline Damianos

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: gr
Re: Rigol dg1022u amplitude issue
« Reply #21 on: December 23, 2017, 08:18:22 pm »
Termination makes sense for the source and the line, not for the load.
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #22 on: December 23, 2017, 08:30:06 pm »
Termination makes sense for the source and the line, not for the load.

A normal 50ohm termination will do the job, ill take it from ebay
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #23 on: December 23, 2017, 08:45:45 pm »
Lets try a little more:
You use a coaxial cable without a proper termination. So, the dominant characteristic is its capacitance plus the inductance and the resistive losses. To verify if this is valid, set the generator to slow sweep from 10MHz to 25MHz and observe if there are "tops and valleys".
Try to find a terminator, for these frequencies even a simple resistor can do the job.
Also verify the compensation of your probe, assuming that you use a 10:1 type, an 1:1 is not proper in this case. Try to do this with a square signal around 2Vp-p and 1MHz. Use it directly at the output of the generator.

So if I hook to scope a bnc to crocodiles, putting a 1Mohm resistor in parallel, and attach another bnc to croco to the F.G and connect to the resistor can be ok?
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #24 on: December 23, 2017, 09:00:49 pm »
Termination makes sense for the source and the line, not for the load.

And the correct way is to use 10X?*
 

Offline Damianos

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: gr
Re: Rigol dg1022u amplitude issue
« Reply #25 on: December 23, 2017, 09:39:25 pm »
If I understand correctly what you mean, no!

If you use a coaxial line (such as RG58 and BNC), you have to terminate it at its end, that is at the oscilloscope input. The resistor must be equal to characteristic impedance of the line, that is 50 Ohms.

The probe (the 10:1) has inside a resistor in series and makes a divider with the input resistance of the oscilloscope. But, because there is also a capacitance there, it has also a trimming capacitor to compensate for that. If the compensation is not correctly adjusted, then the division depends on the frequency. A proper adjustment minimizes capacitive loading.

Avoid to use the grounding wire of the probe, remove it. If the probe has a BNC adaptor, use it. If not, try to attach the probe on the connector of the generator directly; in this case the "spring component" of the probe may help to connect the ground.

Performing the sweep experiment, which I suggested earlier, can find out if the problem is generated by wiring or something else.
 
The following users thanked this post: Samogon, NiaDebesis

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue
« Reply #26 on: December 23, 2017, 09:51:04 pm »
If I understand correctly what you mean, no!

If you use a coaxial line (such as RG58 and BNC), you have to terminate it at its end, that is at the oscilloscope input. The resistor must be equal to characteristic impedance of the line, that is 50 Ohms.

The probe (the 10:1) has inside a resistor in series and makes a divider with the input resistance of the oscilloscope. But, because there is also a capacitance there, it has also a trimming capacitor to compensate for that. If the compensation is not correctly adjusted, then the division depends on the frequency. A proper adjustment minimizes capacitive loading.

Avoid to use the grounding wire of the probe, remove it. If the probe has a BNC adaptor, use it. If not, try to attach the probe on the connector of the generator directly; in this case the "spring component" of the probe may help to connect the ground.

Performing the sweep experiment, which I suggested earlier, can find out if the problem is generated by wiring or something else.

My english is not very well ahah another mistale due to my cheap english..I mean to terminate with a resistor (50 or 1M) at the bnc side with a resistor, and it worked! The problem with the probe was the compensation..that was causing all this problems

I thank you all for the patient with me..but for me there are some language barrior..specially thank you for the suggestion! Compensation..argh..
 

Offline Damianos

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: gr
Re: Rigol dg1022u amplitude issue [solved]
« Reply #27 on: December 23, 2017, 09:58:56 pm »
I'm glad if I helped even a little, but more because there is no problem/damage to any of your devices!
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue [solved]
« Reply #28 on: December 23, 2017, 10:02:18 pm »
I'm glad if I helped even a little, but more because there is no problem/damage to any of your devices!

Again, thank you very much!
 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: Rigol dg1022u amplitude issue [solved]
« Reply #29 on: December 24, 2017, 01:38:24 am »
So much ladened with confused theory, I just wonder how you are going to do troubleshooting field measurements when the point you want to probe at is of "unknown" impedance?

Theories sounded like textbook I must say, yeah!

Practically dumpfounded - How much voltage drop will you encounter if DG1022U is at high impedance, 10Vpp, sine wave at 25Mhz, and the oscilloscope is set to 1Megohm input impedance over the RG58 - BNC to BNC cable of length 2m? [knowing that the characteristic impedance of the RG58 cable is 50 ohms]. ?
 
 |O

With that findings, you build onto your foundation only to move forward.

Cheers;

 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue [solved]
« Reply #30 on: December 24, 2017, 07:46:37 am »
So much ladened with confused theory, I just wonder how you are going to do troubleshooting field measurements when the point you want to probe at is of "unknown" impedance?

Theories sounded like textbook I must say, yeah!

Practically dumpfounded - How much voltage drop will you encounter if DG1022U is at high impedance, 10Vpp, sine wave at 25Mhz, and the oscilloscope is set to 1Megohm input impedance over the RG58 - BNC to BNC cable of length 2m? [knowing that the characteristic impedance of the RG58 cable is 50 ohms]. ?
 
 |O

With that findings, you build onto your foundation only to move forward.

Cheers;

If I have HighZ on the output of the F.G and HigZ on the input of the oscilloscope, they are matched, and there will no drop, cause the F.G adjust it self to a proper level, due to the set of the inpedance
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue [solved]
« Reply #31 on: December 24, 2017, 07:54:00 am »
https://imageshack.us/i/pme051OBj = photo of the connections (I don't have a probe to bnc adapter)

https://imageshack.us/i/pmGhcSVhj = settings on the Rigol dg1022U

https://imageshack.us/i/pmrMaFV7j = scope measuraments



 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: Rigol dg1022u amplitude issue [solved]
« Reply #32 on: December 24, 2017, 10:03:20 am »
BNC to BNC means you do not use the Probe. Connect directly from FG to scope.
You know why Rigol include a BNC to BNC connector cable for you? It's not to make funny terminating resistors, definitely.

From the photo, You still do not have the cable?
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue [solved]
« Reply #33 on: December 24, 2017, 11:09:48 am »
BNC to BNC means you do not use the Probe. Connect directly from FG to scope.
You know why Rigol include a BNC to BNC connector cable for you? It's not to make funny terminating resistors, definitely.

From the photo, You still do not have the cable?

It's incorrect hooking the F.G to the Scope without a termination, that the scope doesn't have
You will have amplitude issue, or phase shift
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue [solved]
« Reply #34 on: December 24, 2017, 11:14:58 am »
And as Damianos says, a 1:1 connection (bnc to bnc) is wrong in this case
 

Offline Damianos

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: gr
Re: Rigol dg1022u amplitude issue [solved]
« Reply #35 on: December 24, 2017, 11:37:21 am »
So much ladened with confused theory, I just wonder how you are going to do troubleshooting field measurements when the point you want to probe at is of "unknown" impedance?

Theories sounded like textbook I must say, yeah!

Practically dumpfounded - How much voltage drop will you encounter if DG1022U is at high impedance, 10Vpp, sine wave at 25Mhz, and the oscilloscope is set to 1Megohm input impedance over the RG58 - BNC to BNC cable of length 2m? [knowing that the characteristic impedance of the RG58 cable is 50 ohms]. ?

 |O

With that findings, you build onto your foundation only to move forward.

Cheers;
It seems that you have confused a little. Here are discussed only simple and practical things.

To answer your question, I will try to explain some simple things:
- The output amplifier, of a typical function generator, has a very low output impedance. For practical reasons there is a resistor connected in series from it to the output connector. This resistor, almost always, has a value of 50 Ohms. This means that when any load is connected at the connector it is created a voltage divider with the upper resistor having this 50 Ohms value. The setting of the load impedance, on the generator, simply adjusts the amplitude of the amplifier to have on the load the "requested" value. It does nothing else.
- To simplify things about the transmission line (the RG58 cable): when it is terminated with a "pure resistance" equal to its characteristic value, then this resistance is "reflected" to the input. Anything else produces disturbances that need a little physics and mathematics to analyze them... (search for forward/reflected/standing waves on transmission lines). Also, have a look at the other specifications of the RG58, such as capacitance/resistance/inductance per length unit.
- It is already described a little the voltage divider of the probe...
- It is also proposed an experiment by sweeping the frequency... (even if the mentioned length of the line seems that is short relative to the wavelength, around 1/8, it can reveal some disturbances)
If you separate each one of the above, you will see that they are not a "mountain" but simple things...

In conclusion: you have to transfer the signal with a transmission line, up to a point and then measure it with a high-impedance probe. In accordance with the above, terminate the line with a suitable resistor and connect the properly adjusted probe on this resistance.
 

Offline NiaDebesisTopic starter

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: it
Re: Rigol dg1022u amplitude issue [solved]
« Reply #36 on: December 24, 2017, 11:53:27 am »
So much ladened with confused theory, I just wonder how you are going to do troubleshooting field measurements when the point you want to probe at is of "unknown" impedance?

Theories sounded like textbook I must say, yeah!

Practically dumpfounded - How much voltage drop will you encounter if DG1022U is at high impedance, 10Vpp, sine wave at 25Mhz, and the oscilloscope is set to 1Megohm input impedance over the RG58 - BNC to BNC cable of length 2m? [knowing that the characteristic impedance of the RG58 cable is 50 ohms]. ?

 |O

With that findings, you build onto your foundation only to move forward.

Cheers;
It seems that you have confused a little. Here are discussed only simple and practical things.

To answer your question, I will try to explain some simple things:
- The output amplifier, of a typical function generator, has a very low output impedance. For practical reasons there is a resistor connected in series from it to the output connector. This resistor, almost always, has a value of 50 Ohms. This means that when any load is connected at the connector it is created a voltage divider with the upper resistor having this 50 Ohms value. The setting of the load impedance, on the generator, simply adjusts the amplitude of the amplifier to have on the load the "requested" value. It does nothing else.
- To simplify things about the transmission line (the RG58 cable): when it is terminated with a "pure resistance" equal to its characteristic value, then this resistance is "reflected" to the input. Anything else produces disturbances that need a little physics and mathematics to analyze them... (search for forward/reflected/standing waves on transmission lines). Also, have a look at the other specifications of the RG58, such as capacitance/resistance/inductance per length unit.
- It is already described a little the voltage divider of the probe...
- It is also proposed an experiment by sweeping the frequency... (even if the mentioned length of the line seems that is short relative to the wavelength, around 1/8, it can reveal some disturbances)
If you separate each one of the above, you will see that they are not a "mountain" but simple things...

In conclusion: you have to transfer the signal with a transmission line, up to a point and then measure it with a high-impedance probe. In accordance with the above, terminate the line with a suitable resistor and connect the properly adjusted probe on this resistance.

Nobody would explain that better than this
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf