Get a load of this... here's what the square wave transition of a Rigol DG1022Z function generator looks like at 1kHz (good) and 1 cycle per 20 sec (bad). Both 6V p/p output. What's going on???
I suspect that is what happens with a DDS type function gen when you "slow it down" to the extreme.
The datasheet says the risetime is <10ns for a 1Vpp output, have you verified that it's able to meet that spec at the very low frequency?
Get a load of this... here's what the square wave transition of a Rigol DG1022Z function generator looks like at 1kHz (good) and 1 cycle per 20 sec (bad). Both 6V p/p output. What's going on???
Can you try this same and show how it looks if you use
Pulse function instead of Square function.
Pulse period 1ms (1kHz) and then 20s (0.05Hz) (and pulse duty 50%)
Same thing with my DG1032Z. It must be a Rigol "Feature".
Mine does this as well on pulse and on square. That's a bit shit!
It's slightly better with an inline terminator at the scope end.
In case anyone cares... Here's the result at 0.05 Hz from a 60 dollar Ming Ho MHS-5200A, connected with a coax and 50 ohm feedthru:
Think I’ve had enough of the bugs in mine and the shitty counter. eBay time!
Rising and falling edge of ramp waveform is good when set to 10nS and 0.02Hz so why is it bad on square wave? How do you report these issues to Rigol? Do they actually take any notice?
Yes, Rigol takes notice. However, it may take many reports to them in order to be investigated. The more owners who report it to Rigol, the better. Unfortunately, they don't seem to monitor the forum (but they should).
If Rigol show this same error also with Pulse function it is really weird. Is it possible these waveforms are produced using Arb memory?
Here is how it looks with (unmodified) SDG1032X using Square wave function.
6Vpp
1mHz (1000s period)
50mHz (20s period)
1Hz
1kHz
I did a test with my Rigol DG1032Z. I looked at Squarewave and Pulse, 6Vpp, 20second period.
No problem at all:
http://www.miedema.dyndns.org/eevblog/Rigol-DG1032Z-20sec-squarewave.gifSame waveform with or without termination. (screen shot is without)
So the strange behaviour is not always a "Rigol feature" ....
Firmware of my DG1032Z is 00.01.12.
Regards, Gertjan.
Gertjan,
I think you need to reduce the timebase to the same 50ns/div as the OP before you can tell if you have the same issue/feature or not. Looking at the edges at 2s/div isn't going to show it.
Hi H.O,
Off coarse you are right....
So I checked again with 50nS timebase:
http://www.miedema.dyndns.org/eevblog/Rigol-DG1032Z-20sec-squarewave-tb=50ns.gifAnd indeed, same behaviour as the other Rigol AWGs....
I guess I never looked very close at the rise times at very low frequencies.....
edit: I measure a "rise time"of 163nS. The Rigol spec is <10nS @ 1Vpp.
So i looked at 1Vpp generator output level as well. Same staircase waveform. So definitely not in spec....
(But without the "staircase" rise time would be in spec)
regards, Gertjan,
This is an artifact of the architecture I think.
The Siglent has “true arb” which is a step beyond a naïve DDS implementation.
The Rigol probably has a square wave lookup table with some sort of slew rate limiting hard coded in. When you play it back at 0.05Hz you get the above.
Rising and falling edge of ramp waveform is good when set to 10nS and 0.02Hz so why is it bad on square wave? How do you report these issues to Rigol? Do they actually take any notice?
Nope, I think its not.
Same problem is there with pulse, right?
For me its not an issue, since I dont use my DG1022Z at very low frequencies. It still should be reported to Rigol, because its a issue nontheless.
If it behaves the same at 1Vpp where the spec says it should be <10ns then it's an issue for sure. I just tested this on my DG4000 and expected the same result but its edges looks the same at 10s period as they do at 10us - for whatever that's worth. On the DG4000 the square waveform is "better" than the pulse waveform when it comes to edge times at lower frequencies.
If it behaves the same at 1Vpp where the spec says it should be <10ns then it's an issue for sure. I just tested this on my DG4000 and expected the same result but its edges looks the same at 10s period as they do at 10us - for whatever that's worth. On the DG4000 the square waveform is "better" than the pulse waveform when it comes to edge times at lower frequencies.
Yes, there is indeed a problem there with Pulse, even at 1Vpp..
Looks like interpolation between waveform values. I bet you see the same thing on any large jump on a waveform at slower frequencies.
Just the usual tradeoffs between performance, samples, frequency, jitters and so on in a digital system. This is most likely the worst case, and on a smoother waveform would most likely be the correct thing to do..
.
Maybe the output is the sum of four ADCs offset by a quarter of a sample period - would explain the pattern.....
Ok, tested it a lot and I think IMHO that the instrument behaves quite right for any frequency =>1Hz
Also seen that the Square function gives better rise time than the Pulse.
Additionally, did set the voltage to max. 20Vpp and the rise time with freq 1Hz was again around 10uS.
For me and its intended usage here its still ok and quite good value for money.
Also seen that the Square function gives better rise time than the Pulse.
Don't forget Pulse function has adjustable rise and fall times.
May not have been set to the minimum of 10nS.
If it behaves the same at 1Vpp where the spec says it should be <10ns then it's an issue for sure. I just tested this on my DG4000 and expected the same result but its edges looks the same at 10s period as they do at 10us - for whatever that's worth. On the DG4000 the square waveform is "better" than the pulse waveform when it comes to edge times at lower frequencies.
Yes, there is indeed a problem there with Pulse, even at 1Vpp..
I will show same with SDG1032X, attached images.
Pulse freq 20mHz 1Vpp
With 50% duty and also set for 50ns width and same 20mHz frequency (50s period)
Just the usual tradeoffs between performance, samples, frequency, jitters and so on in a digital system. This is most likely the worst case, and on a smoother waveform would most likely be the correct thing to do..
This looks more like result of poor desig than "usual tradeoff...etc"
I wonder how there can fine adjust pulse rise and fall times and pulse width with these long period times pulses when transitions are like in these examples.
------------
In my example (in previous msg and now also here) there is just normal simple cheap DDS Arb function generator. (SDG1032X)
In first image pulse period 50s (f=20mHz)
Second image same 50s period but pulse width adjusted for 50ns. (and width and rise/fall times can fine adjust <1ns resolution independent of pulse period, even if period is 1000000s. (factory versions are limited for min 32.6ns width and min 16.8ns rise and fall. This individual unit Pulse rise and fall minimums are mod.)
Square rise and fall is more fast, around under 4ns. (because these can up to 60MHz square)
Rising and falling edge of ramp waveform is good when set to 10nS and 0.02Hz
Nope, I think its not.
Same problem is there with pulse, right?
I'm talking about the ramp waveform, the sawtooth shape one.
The fast edges of those looked okay at very low frequency.
Don't forget Pulse function has adjustable rise and fall times.
May not have been set to the minimum of 10nS.
The shortest rise time in Pulse can be set only to 178.8uS from what I see when Freq. is 0.05Hz, the instrument cannot get this value lower eg to 10uS.
I'm talking about the ramp waveform, the sawtooth shape one.
The fast edges of those looked okay at very low frequency.
Yes, you are right, in sawtooth the rise is excellent with 0.05Hz, even at 20Vpp
Either way, as I said, the instrument is just ok for me.
PS: Actually, in the previous post I made with Pulse setting, I overlooked that the system is actually informing me that the rise time in the 0.05Hz frequency, will be longer, nomatter the voltage setting, so the instrument is actually giving what it says!
[...]
Ah! Your ham callsign has just reminded me of SV1AWE (Bob Koulouris, the son of SV1BT if I remember correctly from those days --but I may be mistaken about the senior callsign), one of my dearest friends and classmate from high school as well as one of the first SV 70cm EME pioneers! Unfortunately, during the last few years he suffers from the Huntington's Chorea disorder; and nowadays I regret I do not have the time I used to have in order to be visiting him as frequently as I used to be doing...
-George