With what you're describing with the slow state update . I think this may use EEPROM to store the settings. EEPROM has a limited write cycle at around a million write cycles depending on temperature conditions. Being conservative with that sort of memory was wise. And who changes settings then immediately turn it off anyway. They should tie the power button to the update storage function. obvious fix
At least they seem to be listening Martin and trying to move forward. I love the big display on this meter and with the bugs ironed out and for this price I see a very nice meter in the future. It shows how much well deserved respect these people have for your opinion.
The AC filter taking a few moments to settle makes perfect sense as it takes time to charge the capacitor... Don't see a problem. They could have used a smaller value cap but I'm sure it was selected for maximum voltages and current. Short of having an array of capacitors selected by range... Just get some patience ;)
Rigol has been pretty good about their change logs and in some cases has been responsive to user feedback albeit taking some time to implement. Thanks for the update, nice to see that at least they're trying! 73 - Dino KL0S
What's the point of having a "soft" power button then? That's when they should be saving the settings. And even if you write to flash memory every single time it shouldn't be a problem if done right. You don't write to the exact same bytes but cycle through the entire flash or EEPROM (wear levelling). Heck, even if you do it "wrong" these chips have 100,000+ to 1,000,000+ write cycle endurances. I don't see how they would need much space to save every option on this meter. 25+ years ago people were saving machine state information when it detected a hard power off and had to save the information before the charge on the PS caps died. Limited time and memory that didn't have the write cycle life that today's chips have yet it worked fine. This has a soft power button giving them all the time they want to write data before turning off. I'm confused why they are having a problem. What are they using the soft power button for then? They're not using it during shutdown and they are not using it to allow a quick start up (the screen should come on instantly) so what is the point? Just to have a high standby current and a pulsing LED?!
Why didn't they write the firmware to save the settings just before it powers down? At least they're not wearing out the flash by saving every control change to it.
+Eric Hospel Hi Eric, I've asked Siglent to respond to your question. I know they gave it to me ahead of time so I think it should be up on their site soon..but I'll let them confirm.
Martin, thanks for this, did you get a chance to asses the temperature measurements ? At the moment it is very inaccurate, ( reads about 4 C high ), using my sdm3055 with a k series thermocouple and 15R1 firmware, and I want to use it for some data logging.
+Martin Read Hi Martin, I did not take the time to properly assess the temperature issue but that's the key issue 15R2 is supposed to correct. If I have the time I'll role back my update and see if I have the same issue.
If it has got a large filter cap, then why is it overshooting when you disconnect and connect the input ? Where is it getting the extra voltage from ? Would you please consider tearing down the multi meter and show where exactly things happening which is causing this issue?
+CZDrMike It could be, but as there is a series resistor it should not reach the highest value and then comes down to the actual. If you see the point when the probes are connected, then it starts by showing some milli volts (due to the incomplete integration period or something), but all of a sudden it goes 10.2436V and then settles down to the value. If it is due to inductance then its a bit difficult to charge the internal cap at that fast rate. I am assuming there is input series resistor which is certain for input protection mechanism. So the inductance factor might not work as per my views. Another point is, when the filter was disabled, then the overshoot didn't occur and the same inductance factor should have worked if it was meant to be the reason. Its interesting fact and I just cant find why its happening. However, if Martin is confident enough to let us see how the inside looks and the input sections are, then it will certainly reveal some of the interesting facts that we might be looking for :)
+Satyajit Roy That's certainly a consideration if I can grab the time. The Signal Path does a look inside in his review so you may spot something there in his video.
With what you're describing with the slow state update . I think this may use EEPROM to store the settings. EEPROM has a limited write cycle at around a million write cycles depending on temperature conditions. Being conservative with that sort of memory was wise. And who changes settings then immediately turn it off anyway. They should tie the power button to the update storage function. obvious fix
Thanks Martin. It is good to see all the deep characterization of this meter and how Siglent has been responding with updates.
At least they seem to be listening Martin and trying to move forward. I love the big display on this meter and with the bugs ironed out and for this price I see a very nice meter in the future. It shows how much well deserved respect these people have for your opinion.
+Gary g1fsh Thanks for the post Gary.
Thanks for the video Martin!
The AC filter taking a few moments to settle makes perfect sense as it takes time to charge the capacitor... Don't see a problem. They could have used a smaller value cap but I'm sure it was selected for maximum voltages and current. Short of having an array of capacitors selected by range... Just get some patience ;)
They should have implemented a digital filter...
Rigol has been pretty good about their change logs and in some cases has been responsive to user feedback albeit taking some time to implement. Thanks for the update, nice to see that at least they're trying! 73 - Dino KL0S
As a resident of Merseyside UK, I just love that t-shirt Martin. By the way, the video's cool too!
In comparison to ANY handheld, this multimeter shows EXACTLY HALF the resistance value measured in circuit.
Why is that ?
They should just save the settings when it soft power-off. But, does it have a way to do reset to defaults if something messed up?
What's the point of having a "soft" power button then? That's when they should be saving the settings. And even if you write to flash memory every single time it shouldn't be a problem if done right. You don't write to the exact same bytes but cycle through the entire flash or EEPROM (wear levelling). Heck, even if you do it "wrong" these chips have 100,000+ to 1,000,000+ write cycle endurances. I don't see how they would need much space to save every option on this meter.
25+ years ago people were saving machine state information when it detected a hard power off and had to save the information before the charge on the PS caps died. Limited time and memory that didn't have the write cycle life that today's chips have yet it worked fine. This has a soft power button giving them all the time they want to write data before turning off. I'm confused why they are having a problem.
What are they using the soft power button for then? They're not using it during shutdown and they are not using it to allow a quick start up (the screen should come on instantly) so what is the point? Just to have a high standby current and a pulsing LED?!
Oh, I see they have "fixed many other problems," according to their release notes. How informative!
Why didn't they write the firmware to save the settings just before it powers down? At least they're not wearing out the flash by saving every control change to it.
+Coolkeys2009 Valid point. I think they will be reading these comments and hopefully they can provide an answer.
Where can I get the new firmware? I don't see it on the Siglent site.
+Eric Hospel Hi Eric, I've asked Siglent to respond to your question. I know they gave it to me ahead of time so I think it should be up on their site soon..but I'll let them confirm.
+mjlorton Thanks Martin. I did get a version of the 15R2 firmware and the problem is fixed!
+Eric Hospel Brilliant, thanks for posting the feedback.
Hi Martin, Did you try (Filter On and Off VDC test) with 10GΩ input impedance setting?
Martin, thanks for this, did you get a chance to asses the temperature measurements ? At the moment it is very inaccurate, ( reads about 4 C high ), using my sdm3055 with a k series thermocouple and 15R1 firmware, and I want to use it for some data logging.
+Martin Read Hi Martin, I did not take the time to properly assess the temperature issue but that's the key issue 15R2 is supposed to correct. If I have the time I'll role back my update and see if I have the same issue.
If it has got a large filter cap, then why is it overshooting when you disconnect and connect the input ? Where is it getting the extra voltage from ? Would you please consider tearing down the multi meter and show where exactly things happening which is causing this issue?
+Satyajit Roy I think that perhaps this overshoot has to do something with inductance of wires (just my theory)
+CZDrMike It could be, but as there is a series resistor it should not reach the highest value and then comes down to the actual. If you see the point when the probes are connected, then it starts by showing some milli volts (due to the incomplete integration period or something), but all of a sudden it goes 10.2436V and then settles down to the value. If it is due to inductance then its a bit difficult to charge the internal cap at that fast rate. I am assuming there is input series resistor which is certain for input protection mechanism. So the inductance factor might not work as per my views. Another point is, when the filter was disabled, then the overshoot didn't occur and the same inductance factor should have worked if it was meant to be the reason. Its interesting fact and I just cant find why its happening. However, if Martin is confident enough to let us see how the inside looks and the input sections are, then it will certainly reveal some of the interesting facts that we might be looking for :)
+Satyajit Roy That's certainly a consideration if I can grab the time. The Signal Path does a look inside in his review so you may spot something there in his video.
SD "M" not SD :P
can u measire less than 1 milli ohm
China #1
You should hack it