-
Haven't seen anyone reporting this particular version of the issue, so here goes:
On this DS1104Z-S (00.04.01.SP2/0.1.1), using a 10MHz square wave from the internal sig gen, I see jitter at 5us, but it's very slight, it's about 800ps vs. 400ps at 0us. I tried different signals and frequencies, and I don't see any correlation between rise time and jitter amount.
AC Trigger issue is there. Wouldn't know what to use AC triggering for though
Triggering on both flanks issue: I get the same behaviour with some choices of hold-off, even at values that are smaller than half a period. I'm getting the impression that the real holf-off time has some dead time added to it. Changing the value a bit quickly gets a proper eye pattern on screen in all trigger modes though. -
Ok, so this only seams to be a problem if your DC coupling and have a DC offset and you don't want to have to adjust the trigger point when the offset changes. It also occurs if you for some reason want to AC couple and AC trigger at the same time..not sure why.
In the end this would impact a very small percentage of users and in many cases can be worked around.
Has anyone tried using the External trigger in DC coupled mode yet with just a basic input cap to make it in effect AC coupled ?
Jeff -
Ok, to follow up on my own suggestion I tried using the external trigger on my DS2000A with a capacitor input on the trigger to effectively AC couple the trigger. Works like a charm. Granted this is hardly optimal..it might be an option for some however that need that AC coupled trigger. Interestingly it appears the external trigger is always DC coupled, changing it to AC doesn't seam to have any effect. My little capacitor trick gets around that as well.
Just my two cents.
Jeff -
Why the user should be doing this. The fact that the issue has been known for two years and Rigol did not bother fixing it and even not removed/hide that option in the menu to me is a indication that Rigol is still a Mickey Mouse company. If i have a instrument on my bench i want to use it and trust it, not fight with it.
-
Why the user should be doing this. The fact that the issue has been known for two years and Rigol did not bother fixing it and even not removed/hide that option in the menu to me is a indication that Rigol is still a Mickey Mouse company. If i have a instrument on my bench i want to use it and trust it, not fight with it.
Just my personal Opinion here, but it seams the Chinese companies can produce some really good hardware at times, their software in many cases just seams to be lacking. I think Rigol is the better of the bunch in that area. I've seen so many other cases of a great Chinese product crippled by their firmware (software). The English speaking based companies seam to do a bit better job with the firmware. Rigol should hire a bunch of Russians to write all their code, that's one group that can make make next to nothing do just about anything.
Just my opinion.
Jeff -
The Rigol firmware seems pretty good to me.
I'm not saying the user interface can't be improved (especially the use of the multi-function knob!!) but there's some complex code in there and it seems to hang together quite well.
-
Hello everyone. I am not sure if you saw the Rigol response back. If not here it is:
https://www.tequipment.net/assets/1/7/EEVBlog_message1.pdf
and wtf is inside? a survey link from dumb MBA ...
let's answer some of these stupid questions, coming from an "Director of Product Marketing & SW Applications". Ohhh yeah.
- How often do you use AC coupling on your scope?
every fucking day
- When you use AC coupling on a channel, why do you do it?
not your fucking business, and btw, wtf you mean with this question?
So you trying to tell me that next time when i buy car i will get no reverse gear when i aswer in survey "i never use reverse gear" I paid for function introduced in datasheet/user manual/product description and you guys asking such stupid questions? Fix that shit in the fw or replace broken mainbaords, and stop asking such stupid things. Every other manufacturer (and even half of the cheap DIY china scopes) is capable of implementing AC coupled trigger/channel, so where is the problem for Rigol? This is very basic function, when you don't know this (and yeah, can you please remove all these shitty "skils" from your LinkedIn profile? this is pain in the ass, you have no idea about most of them, so please, remove them), then i think you should really think about new job. You worked before for Keithley and now Rigol and you have no idea about such simple things? -
Why the user should be doing these tricks and workarounds The fact that the issue has been known for two years and Rigol did not bother fixing it and even not removed that option in the menu to me is a indication that Rigol is still a Mickey Mouse company. If i have a instrument on my bench i want to use it and trust it, not fight with it.
-
So you trying to tell me that next time when i buy car i will get no reverse gear when i aswer in survey "i never use reverse gear" I paid for function introduced in datasheet/user manual/product description and you guys asking such stupid questions?
Bad analogy. It's not as fundamentally broken as "no reverse gear".
It's more like "the number plate light isn't as bright as I normally like it".Fix that shit in the fw or replace broken mainbaords, and stop asking such stupid things. Every other manufacturer (and even half of the cheap DIY china scopes) is capable of implementing AC coupled trigger/channel, so where is the problem for Rigol?
Rigol is making 4-channel, 100MHz 'scopes and selling them for $400.
Your rant just sounds like you bought a half price horse and are complaining about the teeth. I'm sure most countries have a saying about that.Rigol is still a Mickey Mouse company. If i have a instrument on my bench i want to use it and trust it, not fight with it.
Feel free to spend a lot more money...it's your perogative.
-
Feel free to spend a lot more money...it's your perogative.
I spent $3900 on a 4000 series, been fighting with Rigol for almost a year to get issues fixed, they don't care. -
Feel free to spend a lot more money...it's your perogative.
I spent $3900 on a 4000 series, been fighting with Rigol for almost a year to get issues fixed, they don't care.
Hmmm. OK, that's not so good. They should look after their $4000 customers...
-
Rigol is making 4-channel, 100MHz 'scopes and selling them for $400.
Your rant just sounds like you bought a half price horse and are complaining about the teeth. I'm sure most countries have a saying about that.
there must be something wrong with you:
1- in lot of countries one have to give lot of blowjobs to earn $400 (and cheap does not means bad or broken)
2- fully loaded DS2000A costs nearly $4000, and it does have same shit bug (that should be enought to care, right?)
Rant? how can an serious manufacturer ask such stupid questions? (except Q3, i have no idea why ppl are complained about, on TEK this was never ever working and nobody was crying about, it does not even make any sense to be implemented. But Q1 and Q2 are "US of Idiocracy-like" questions, from an MBA with RF and Electronic my ass skills, there is no hope for the human race). -
In such situations I do very simple. Disassemble device in a way it cannot be noticed. It might be tricky to remove Rigol sticker but I've done that because of the crappy service when they refuse to admit an obvious fault. Then kill an electronics. Have done that by destroying particular IC or just changing some resistor value so I can simply fix back if something goes wrong. In Rigol case to not remove security sticker, might just stick wire inside and apply some voltage without disassembly at all. Bring them completely malfunctioning deviceFeel free to spend a lot more money...it's your perogative.
I spent $3900 on a 4000 series, been fighting with Rigol for almost a year to get issues fixed, they don't care.
. Don't feel any remorse as this is not your fault that somebody is a jerk, and you just made things to be done as they should be in the first place. 
-
In such situations I do very simple.
I know many people do such a thing and feel justified because they feel screwed or whatever. My personal belief, right or wrong, is to just be honest.
When I retired from the mega corporation to work self employed, I thought I could easily ignore minor nuances and benefit from the value Rigol provides. Well, lets just way, boy did I underestimate the nuances!
Anyhow, I've learned my lesson with Rigol. With companies like Hameg and Agilent (and probably Tek, just don't have a bunch of personal experience with them), when you have a legitimate issue, it gets resolved in adequate time. -
My personal belief, right or wrong, is to just be honest.
I'm honest on the first try. So I'm honest until another party is honest too. -
I'm honest on the first try. So I'm honest until another party is honest too.
Honesty and morals don't take a day off. Just because someone else or some company decides to lie and deceive never is an excuse to do the same. -
If nobody was going to repair faulty PCB anyway as it often happens nowadays, this does not change anything for another party. PCB was faulty, it just became more faulty so they cannot get away one more time. This is much more honest rather than overclocking computer hardware to the hell and then request warranty. Or destroying mobile phone firmware by flashing modded one and requesting warranty too.
I'm honest on the first try. So I'm honest until another party is honest too.
Honesty and morals don't take a day off. Just because someone else or some company decides to lie and deceive never is an excuse to do the same.
I'm not in US, you know, where usually you get replacement without questions, even if there was no fault at all. Funny thing that you pay 1.5x US price and must prove that you are not elephant when device breaks. -
- How often do you use AC coupling on your scope?
every fucking day
This basically sums it up for me. Most of the oscilloscopes I use (old ones) support peak-to-peak or auto level triggering so they do not need AC coupled triggering although most support both. The ones which lack peak-to-peak and auto level triggering default to AC coupled triggering which is almost as good and sometimes better. DC coupled triggering is needed at low input frequencies and for single shot and sometimes normal timebase mode so you always need to have it available.
If someone has never used an oscilloscope with peak-to-peak or auto level triggering, I could see not missing them but I consider one of these to be a must have feature since before 1970. I did not realize the Rigols lacked both until now but this just makes AC coupled triggering more important. Can you get by without it? Sure, but it makes the oscilloscope more difficult to use.Quote- When you use AC coupling on a channel, why do you do it?
not your fucking business, and btw, wtf you mean with this question?
So I do not have to mess with the trigger level when probing different signals and using different offsets.So you trying to tell me that next time when i buy car i will get no reverse gear when i aswer in survey "i never use reverse gear" I paid for function introduced in datasheet/user manual/product description and you guys asking such stupid questions?
Bad analogy. It's not as fundamentally broken as "no reverse gear".
It's more like "the number plate light isn't as bright as I normally like it".
I think tinhead's analogy is pretty good. How often do you use reverse gear? You can always park where you do not need reverse or get out and push your car backwards.
What occurred to me was the Pentium FDIV bug as related by Bob Colwell (chief IA-32 architect on the Pentium Pro, Pentium II, Pentium III, and Pentium 4) where someone asked Andy Grove if he would accept a new car with has a big scratch on the driver's door. The gas mileage is the same and the cargo capacity is the same and the radio still works so what is not to like? Are you going to take it home? Hell no! You fix that because I paid for a car in perfect condition.
Public: I paid for a fully working chip, so give me one.
Intel: Nothing is bug-free, this bug doesn't affect you.
Public: I will judge the severity. I won't trust you to do it.
Intel: We've always done it before. You can't anyhow.
Public: Stand behind your brand or suffer the consequences.
Intel: Ok. We give up. Recall. $475M (ouch!)
I am inclined to think Rigol cannot solve the AC coupled triggering bug since it has been going on for 2 years in the DS2000A series and the 5us jitter bug seems like it would be even more intractable. Maybe they have their top men working on it. Who? Top ... men.
-
I don't get why the LF Reject improves the stability of the Jitter, LF Reject removes the lower frequencies (< ~20khz?). This negates IMO any concept that it's a sample rate issue or such. It's almost like some stray signal is getting in to the trigger circuit or perhaps a DC offset on some op-amp or something. IDK. It's just weird.
Jeff -
I don't get why the LF Reject improves the stability of the Jitter, LF Reject removes the lower frequencies (< ~20khz?). This negates IMO any concept that it's a sample rate issue or such. It's almost like some stray signal is getting in to the trigger circuit or perhaps a DC offset on some op-amp or something. IDK. It's just weird.
AC trigger coupling is just low frequency reject with a lower cutoff frequency so it works below 50/60 Hz. With such a low cutoff frequency, if the filter (DSP in the FPGA?) is reset after every trigger, then it may not have enough time to settle before the new trigger occurs. The low frequency reject trigger settles much more quickly.
You can sometimes see this happen on analog oscilloscopes which support both; in AC coupled trigger mode, the trigger position will initially visibly shift horizontally while in low frequency rejection mode, it settles much faster. A 30 Hz low frequency cutoff has a time constant of 11.7 milliseconds while a 30 kHz low frequency cutoff has a time constant of 11.7 microseconds.
-
I don't get why the LF Reject improves the stability of the Jitter, LF Reject removes the lower frequencies (< ~20khz?). This negates IMO any concept that it's a sample rate issue or such. It's almost like some stray signal is getting in to the trigger circuit or perhaps a DC offset on some op-amp or something. IDK. It's just weird.
AC trigger coupling is just low frequency reject with a lower cutoff frequency so it works below 50/60 Hz. With such a low cutoff frequency, if the filter (DSP in the FPGA?) is reset after every trigger, then it may not have enough time to settle before the new trigger occurs. The low frequency reject trigger settles much more quickly.
You can sometimes see this happen on analog oscilloscopes which support both; in AC coupled trigger mode, the trigger position will initially visibly shift horizontally while in low frequency rejection mode, it settles much faster. A 30 Hz low frequency cutoff has a time constant of 11.7 milliseconds while a 30 kHz low frequency cutoff has a time constant of 11.7 microseconds.
I have a 2465B in mint condition. Has the same LF and HF rejects. I was under the impression the LF reject removed low frequencies from the trigger path while HF Reject did the opposite (i.e. removed high frequencies). So still it seams odd the LF Reject stabilizes the jutter to a degree while the HF Reject does not. Makes me think their is a low freq element that's impacting things a bit. Naturally my Tek doesn't suffer from these issues with the same signals.
Jeff -
So, it seems the sample clock *is* spread. Maybe not on purpose? It's difficult to unwind if you're trying to get a decent FFT out of the thing.
Below are a couple shots from an SA looking at the ADC (HMCAD1511) CLKP and CLKN inputs with a differential probe. These inputs are being generated by an ADF4360-7 frequency synthesizer, which has as its base clock a 25MHz crystal/resonator/oscillator block (not sure exactly, marked "CETDCJ 25.000").
I'm noting that the synthesizer output has discrete jumps of 100kHz. Now where have we seen that number before??
The 5us jitter is pretty bad on this unit.
Still trying to glue all the observations together, but thought I'd post since it's interesting.
ADC: http://www.hittite.com/products/view.html/view/HMCAD1511
Synth: http://www.analog.com/en/rfif-components/pll-synthesizersvcos/adf4360-7/products/product.html
(Pardon the screen photos - my screen capture utility is not working...)
Edit: firmware 4.02.SP3, board vers 0.1.1
-
AC trigger coupling is just low frequency reject with a lower cutoff frequency so it works below 50/60 Hz. With such a low cutoff frequency, if the filter (DSP in the FPGA?) is reset after every trigger, then it may not have enough time to settle before the new trigger occurs. The low frequency reject trigger settles much more quickly.
You can sometimes see this happen on analog oscilloscopes which support both; in AC coupled trigger mode, the trigger position will initially visibly shift horizontally while in low frequency rejection mode, it settles much faster. A 30 Hz low frequency cutoff has a time constant of 11.7 milliseconds while a 30 kHz low frequency cutoff has a time constant of 11.7 microseconds.
I have a 2465B in mint condition. Has the same LF and HF rejects. I was under the impression the LF reject removed low frequencies from the trigger path while HF Reject did the opposite (i.e. removed high frequencies). So still it seams odd the LF Reject stabilizes the jutter to a degree while the HF Reject does not. Makes me think their is a low freq element that's impacting things a bit. Naturally my Tek doesn't suffer from these issues with the same signals.
Low frequency reject is just another way to say high pass filter and high frequency reject is just another way to saw low pass filter. The difference between AC coupling and low frequency reject is just the cutoff frequency.
Triggers work best on fast edges which will produce the lowest jitter but sometimes you want to deliberately ignore high frequencies or low frequencies. For example triggering on the AC line will usually be improved with high frequency reject but if you want to trigger on a high frequency glitch which is on the AC line, low frequency reject would be appropriate.
-
It's more like "the number plate light isn't as bright as I normally like it".
i like this.
and i would like to add that i wouldn't even remember it on every day of the week and the cops wouldn't notice it either!
tinheadQuoteEvery other manufacturer (and even half of the cheap DIY china scopes) is capable of implementing AC coupled trigger/channel, so where is the problem for Rigol?
oh, come on!
i think we all read the examples where chinese scopes perform (or not) based on the serial number they have....and not in function that are rarely (if ever) used as this one.
with these bugs it's still best bang for the buck in market today.
or will you care to name something better in that price range?
it can have the same bugs as this one!
ehm...
d.hessQuoteI think tinhead's analogy is pretty good. How often do you use reverse gear? You can always park where you do not need reverse or get out and push your car backwards.
no, it's not good beacuse it's not practical.
imagine sales of a car with reverse gear that is just awfuly noisy.
no, you can't compare even that with a function of a scope, because scopes today have many funcitons many people will rarely use.
most peopla here didn't know what ac trigger is, and that's a fact.
most car drivers know what is reverse and use it every day. it's the thing that prevents them from pushing the car.
otoh i doubt they know all the functions of ther new "entertainment systems" in cars....
this would be like missing an option to play...say....flv files in cars....or an .avi with indeo4 if we wanna be really obscure...i mean most people don't know about these things, so their chances of objecting it are rather diminished.
if you object (ie tinhead) it that much, fix it yourself!
(ie the approach markl just did with trying to figure it out himself)
and then charge to all rigol users that want it fixed, but i predict you'll soon be out of business if you would rely just on that source of income!
this is not to say rigol shouldn't fix it....but somehow i doubt most will would be sending it away to fix this problem, esp. if rigol wouldn't pay the return shipping too.
(i must admit that i dunno would i even change the firmware if it would be fixable in that way...i mean...would i lose the hack options by doing that?
)
to that extent was there ever such a case that electronics company did that, i mean to repair a big amount of units and pay for their shipping?
but i like it how rigol got our hopes up when they said to do self calibration.... heh...
i would say rigol should investigate this and explain it thoroughly, and explain possible fix scenarios....that would be enough from them...not a pdf with few questions...
if you just send it to them for repair, you won't learn anything about your scope!
-
AC trigger coupling is just low frequency reject with a lower cutoff frequency so it works below 50/60 Hz. With such a low cutoff frequency, if the filter (DSP in the FPGA?) is reset after every trigger, then it may not have enough time to settle before the new trigger occurs. The low frequency reject trigger settles much more quickly.
You can sometimes see this happen on analog oscilloscopes which support both; in AC coupled trigger mode, the trigger position will initially visibly shift horizontally while in low frequency rejection mode, it settles much faster. A 30 Hz low frequency cutoff has a time constant of 11.7 milliseconds while a 30 kHz low frequency cutoff has a time constant of 11.7 microseconds.
I have a 2465B in mint condition. Has the same LF and HF rejects. I was under the impression the LF reject removed low frequencies from the trigger path while HF Reject did the opposite (i.e. removed high frequencies). So still it seams odd the LF Reject stabilizes the jutter to a degree while the HF Reject does not. Makes me think their is a low freq element that's impacting things a bit. Naturally my Tek doesn't suffer from these issues with the same signals.
Low frequency reject is just another way to say high pass filter and high frequency reject is just another way to saw low pass filter. The difference between AC coupling and low frequency reject is just the cutoff frequency.
Triggers work best on fast edges which will produce the lowest jitter but sometimes you want to deliberately ignore high frequencies or low frequencies. For example triggering on the AC line will usually be improved with high frequency reject but if you want to trigger on a high frequency glitch which is on the AC line, low frequency reject would be appropriate.
I get all that, I just wonder what additional low freq is getting adding in to the trigger circuit perhaps internally. My 2465B doesn't have any issues with jitter on the same exact signal with the same cable, same impedance..etc..etc. So I have to presume the 2000A I have is getting some other noise in the trigger circuit at some point. I also notice the problem doesn't seam to exist on the External trigger, at least on mine. Then again I can't get it to AC trigger with the external connection without a cap inline.
Jeff