Counts are easy. As for the first digit in the digits notation: 1/2 digit: It doesn't mean "half a digit" but refers to the maximum possible value (1) out of 2 values (0,1). The term "half" has probably become popular for simplicity, but some even use ".5," which is totally misleading. 3/4 digit: This means the meter can display a maximum value of 3 out of 4 possible values (0,1,2,3) 5/6 digit: Here, the maximum display value is 5 out of 6 possible values (0,1,2,3,4,5) In cases where a multimeter shows "2" in the first digit due to overrange, it should be specified in the documentation. The overrange capability isn't always universal for every multimeter, and that's why a "4 1/2 digit" meter can have a count of 22000 (19999) with a 10% overrange but is not called "4 and 2/3 digit." Sometimes there is no overrange at all.
you said, " 3/4 digit: This means the meter can display a maximum value of 3 out of 4 possible values (0,1,2,3) " You mean it could display 0 or 1 or 2 or 3? It's a bit confusing... IMO... to say it " 3/4 digit: This means the meter can display a maximum value of 3 out of 4 possible values (0,1,2,3) "
Great education! Confusing topic and this video really clarified the counts vs. digits in my mind. I have a uni-t with 60000 counts . It's almost too much for my normal use. I rarely use it.
High resolution is very useful if you monitor current through a low resistance shunt and want to see how the current behaves (so if it goes up or down). The last few digits might not be very accurate in an absolute sense, but they are usually stable, so observing if the current is gaining or decreasing at certain instances is accurate. I use the 500k count mode on my Brymen BM869S for this.
Needless to say, this is about display resolution, not actual accuracy of the meter. Those two are often mixed up in conversation or reason to buy a certain meter.
Some of the Siglent SDM meters used to be 240000 or 2400000 counts full scale but at some point the firmware changed and it dropped to 220000 or 2200000 counts full scale.
Great video Kerry - I've been wondering if I should get a 6 1/2 from one of the current lower cost manufacturers seeing that I have a 3458A, 34401A, Keithley 2401 - Do you think they're worth picking up? I do think that I should have got a Rigol 3068 instead of grabbing the 3058 - Thoughts?
IMO nothing can beat an HP 3458A. I think one of the main benefits of a modern 6 1/2 digit meter is some of the more advanced functions such as histogram and other statistical capabilities. But at the end of the day, it really depends on what you need. For my electronics work, I rarely needed the resolution of a 6 1/2 meter and when I did, my old Keithley 196 would usually do just fine. I haven't used the Rigol 3068 but the specs looked promising.
@@KerryWongBlog That is more or less what I was wondering about - Should I put up the coin to buy a modern 6 1/2 - I see videos showing a much more expressive display and think I should get something that acts like that. Anyway - Would appreciate your thoughts on the Uni-T/Rigol/Siglent vs Keysight/Etc choice? Basically, I don't want to over pay and yet I want all the goodies - Damn my consumer basis...
@@tgsoapbox Sounds like you are addicted to DDMs, just like me. I'd say just remind yourself that you are addicted and resist the urge to purchase another one. The first step is admitting you have a problem.
These meters have all have different counts; 22,000, 40,000, 20,0000, and 2,000,000. Why do they do this? Why don't they use all the values of 0 to 9 for each digit, meaning a 5 digit meter could display from 0 to 99,999. I would guess it has something to do with how the A/D works. Can you explain this?
It definitely depends on the ADC and the Vref. If the reference voltage is 2.5V then that would be the maximum signal level the ADC can convert. The linearity of ADC (INL/DNL) also plays a role and designers could limit the range further to minimize non-linearity induced errors.
Counts are easy. As for the first digit in the digits notation:
1/2 digit: It doesn't mean "half a digit" but refers to the maximum possible value (1) out of 2 values (0,1). The term "half" has probably become popular for simplicity, but some even use ".5," which is totally misleading.
3/4 digit: This means the meter can display a maximum value of 3 out of 4 possible values (0,1,2,3)
5/6 digit: Here, the maximum display value is 5 out of 6 possible values (0,1,2,3,4,5)
In cases where a multimeter shows "2" in the first digit due to overrange, it should be specified in the documentation. The overrange capability isn't always universal for every multimeter, and that's why a "4 1/2 digit" meter can have a count of 22000 (19999) with a 10% overrange but is not called "4 and 2/3 digit." Sometimes there is no overrange at all.
you said, " 3/4 digit: This means the meter can display a maximum value of 3 out of 4 possible values (0,1,2,3) " You mean it could display 0 or 1 or 2 or 3? It's a bit confusing... IMO... to say it " 3/4 digit: This means the meter can display a maximum value of 3 out of 4 possible values (0,1,2,3) "
Great explanation! Thanks!
Great education! Confusing topic and this video really clarified the counts vs. digits in my mind. I have a uni-t with 60000 counts . It's almost too much for my normal use. I rarely use it.
High resolution is very useful if you monitor current through a low resistance shunt and want to see how the current behaves (so if it goes up or down). The last few digits might not be very accurate in an absolute sense, but they are usually stable, so observing if the current is gaining or decreasing at certain instances is accurate. I use the 500k count mode on my Brymen BM869S for this.
Nice clear concise explanation 👍👍
Good explanation, thanks Kerry.
شكرا على الفيديو
Interesting info, thanks 👍
Needless to say, this is about display resolution, not actual accuracy of the meter. Those two are often mixed up in conversation or reason to buy a certain meter.
Awesome reminder !
Some of the Siglent SDM meters used to be 240000 or 2400000 counts full scale but at some point the firmware changed and it dropped to 220000 or 2200000 counts full scale.
Thanks Kerry
Great video. The L/R audio bias towards left, up until 2:22, hurts my brain however. Maybe you are bump up the right gain in post or get a lapel mic?
Hi what about fluke 115??
Great video Kerry - I've been wondering if I should get a 6 1/2 from one of the current lower cost manufacturers seeing that I have a 3458A, 34401A, Keithley 2401 - Do you think they're worth picking up? I do think that I should have got a Rigol 3068 instead of grabbing the 3058 - Thoughts?
IMO nothing can beat an HP 3458A. I think one of the main benefits of a modern 6 1/2 digit meter is some of the more advanced functions such as histogram and other statistical capabilities. But at the end of the day, it really depends on what you need. For my electronics work, I rarely needed the resolution of a 6 1/2 meter and when I did, my old Keithley 196 would usually do just fine. I haven't used the Rigol 3068 but the specs looked promising.
@@KerryWongBlog That is more or less what I was wondering about - Should I put up the coin to buy a modern 6 1/2 - I see videos showing a much more expressive display and think I should get something that acts like that.
Anyway - Would appreciate your thoughts on the Uni-T/Rigol/Siglent vs Keysight/Etc choice?
Basically, I don't want to over pay and yet I want all the goodies - Damn my consumer basis...
Kerry - FYI I think I have a gap in my DMMs - Very high end, and low end ones - Just thinking about filling that - If this makes sense
@@tgsoapbox Sounds like you are addicted to DDMs, just like me. I'd say just remind yourself that you are addicted and resist the urge to purchase another one. The first step is admitting you have a problem.
@@tgsoapbox for what purpose?
Ut117c vs fluke 117c can compare dmm
These meters have all have different counts; 22,000, 40,000, 20,0000, and 2,000,000. Why do they do this?
Why don't they use all the values of 0 to 9 for each digit, meaning a 5 digit meter could display from 0 to 99,999.
I would guess it has something to do with how the A/D works. Can you explain this?
It definitely depends on the ADC and the Vref. If the reference voltage is 2.5V then that would be the maximum signal level the ADC can convert. The linearity of ADC (INL/DNL) also plays a role and designers could limit the range further to minimize non-linearity induced errors.
Enlightening!
owon cms101 test plz
I am planning to review that in December, stay tuned!
First! 🙂
All digits must be up to 9. (9999 counts for example)