There is new member in Siglent DMM products.

New pruct is in X-instruments category.

SDM3045

**X**.

Overall functionally like SDM3055 but new SDM3045

**X** is 60000 count, advertised as 4½ digit DMM

- or is it better say 4 3/4 or 4 4/5 or even 4 5/6 as some manufacturer name 60000 count meter.. ( SDM3055 have 240000 and it is named as 5 1/2 digit)

If look what is Tektronix

*opinion*, yes world is free for opinions:

http://www.tek.com/document/whitepaper/understanding-handheld-dmm-specificationsthen it can name least as 4 4/5 digit or more.

But if use other approach to digits:

Resolution digits Digits = log10 (counts) and this give with 60000 count roughly 4 3/4.

There is many meter what say 5 1/2 digit but markets have adopted and accepted these 200000 (199999) counts can name 5 1/2 digits, and also this is well explained. Confusion and different opinions for marketing "digits" start just after this... 1/3, 2/3, 3/4, 4/5, 4/6 and so on.....

There is one road away from this. Talk about counts as primary when talk about resolution and secondary can tell how many digits if even need.

Data sheet pdfResolution is lower but relative to resolution difference true accuracy do not look bad, is it even better in some cases.

(SDM3055 one year DCV is listed as 0.015% and SDM3045

**X** as 0.01%)

If look more carefully accuracy specifications, it need look percent + count total.

If measure 5V and it show exactly 5.0000V (full range 6V) error can be 0.01%(from display) + 6 count.

(one count in this range is 0.0001V)

max(%) 5.0005V + 6 count so max is 5.0011V (0.022% more)

min(%) 4.9995V - 6 count so min is 4.9989V (0.022% less)

But wait a minute, what happend if I measure and display show exactly 10.000V

Now meter is in 60V range. But, in this range specifications tell: 0.02%(from display) + 4 count.

(one count in this range is 0.001)

max(%) 10.002V + 4 count so max is 10.006V (0.06% more)

min(%) 9.998V - 4 count so min is 9.994V (0.06% less)

If these errors specified by number of counts are true random noise without including other type of errors then average can do some trick.

If you wonder these real % take your DMM specifications, read carefully and do calculations.

Also this was just example and not at all like worst case finding. Worst cases are when you look just some border values.

Example just >6V or just >600mV. These is not Siglent special, this kind of things are normal, independent of if meter is Fluke, Keithley, AVO, HuiHai or Utrend.