Review: UNI-T UT61E digital multimeter

What was the price of that clamp?


The manual explicit states that it has no backlight (and the switch does not work either).

HKJ, you made my day. Thanks

If memory serves, about $4000. It was added to the lab a long time ago though, so I might be off by a bit - my memory for unimportant things isn't very long..

PPtk

Nice review! I may have to grab one of these. Didn't know these were so darn cheap! I've got a pricey Agilent datalogging DMM, but could use one for basic voltage logging.

It's funny that the unit you reviewed seems more accurate than the more expensive UNI-T-UT70B (although this model has the temperature feature).

Kreisler why do you have to use multiple usernames on BLF?

I liked your first reply more. Wink

I ordered the UT-203 clamp meter from dealextreme. It is okay, but the zero is way off - about 1.5 amps. It has a 'rel' button than can be abused for zeroing however.

The zero is not really stable but it is possible to figure out tens of mA on it. Compared to a rather anonymous brand at work the zero offset is a bit better on that one (half an amp) but still not perfect.

The Uni-T readings match my normal multimeter pretty well though.

I posted that post after researching the price on eBay and a few other sites. I see $62 Shipped on eBay is the lowest price, but I would probably buy from a US seller @ $69.99.

i did 2 amperage measurement sessions (tailcap readings: 10440 on Hi, 10440 on Med) and saved the 2 XLS files, then opened them with Excel2010, deleted the unnecessary columns (DC/AC, Unit, AUTO) and some rows (beginning and end measurements). i copied the Med-amperage (see "D") measurements to the spreadsheet of the Hi-amperage measurements (see "C"), and finally selected both columns and let Excel2010 produce a line graph:

the 2 line graphs look nice (and we can even see the down-step from Hi to Hi-2 after 7mins; f e n i x l d 0 1 r 4 : unregulated driver with 10440's so the down-step must be a time-controlled feature!!) but the x-axis looks wrong:

  • it should begin with 0.0 (unit time: seconds), and not with "1" as shown
  • the time difference between 2 neighboring data points is 0.5sec (the runtime of 10440-Med is ~30mins, and not the double "3601" as shown)
  • the ticks should be 1 tick per 60 seconds (=1 tick per minute)

I will also try to visualize the data with other graphing software (Wolfram, Origin) but if someone is proficient in Excel and knows how to quickly get my x-axis "corrected", then please shoot me. Obviously i am not proficient in using Excel at all

Thanks for some concrete Excel2007/2010 steps ..

The usual curve in Excel is not a x-y chart, but each value fills one position. If you have a entry at regular time intervals (Like each second), you can just calculate a column with time in second or minutes from start and the use it for "Category" labels. You will also need to edit the axis and define how often to show a value (Like only show 1 in 30 values). This is the method I uses for my runtime charts in my flashlight reviews.

The other solution is to select a real x-y chart, you have to open the full chart type selection to find it (It is called something with X Y). The again you need to calculate a time column and use it as the X value, but this time you do not need regular time intervals.

Hi HKJ, thanks for the tip with X-Y charts. Will try to figure it out now with my Excel2010. At the same time i am downloading Origin. It's easy to generate a 40mins time column with Wolfram but it's not installed on my system.

Looks like Excel isnt the most helpful here..

EDIT: another problem in Excel (or in combination with other software) is the interpretation of comma "," and decimal point ".", argh!!

EDIT2: X-Y charts: Excel returns an error because of some number of rows limitation (255 rows) or freezes. d*mn it. Excel is not recommendable at all if one has to manipulate or visualize thousands of data points!! :(

Excel has a auto fill function, you just enter the first few values, select the area and the uses the auto fill to generate the remaining values.

The point/comma issue is a PITA.

In my runtime charts I sample each second, this means that I can have lots of entries. This works fine with the line chart.

Holy Mother! HKJ

A review on a DMM? Only on BLF :-)

Thank you for taking the time and making the effort to do this. That is so awesome.

I NEVER read a DMM review before although I've seen them. I read this though :-)

Actually sounds a like its a lot of functionality in a cheap meter.

And pilotptk: Good one on the clamp meter.

I figured it out with Excel2010, the UT61E graph is presented in the F e n i x post. Thanks again HKJ for the support, as always!!

to whom it may concern, recalibration of the UT61E isnt necessary imho but it is possible. FOI, some users buy a little product called DMMCheck and do the adjustments on their own.

@HKJ or anyone to whom it may concern for the RS232 converters issue

Nowdays serial is obsolete,so connecting to Pc is not possible for those who does not has a serial port,but I found some RS232 convetor to USB in Deal Extreme,many consumer reported that some converters are not working,based upon my observation I found that the better converters are those which has FTDI chips and not prolific chips,since I am not that technically sound plz suggest which will work for Unit 61E,I have chosen these in Deal Extreme based on the customer reviews which are as follows.

1>http://dx.com/p/usb-2-0-to-rs232-adapter-0-8-meter-9656
(highly praised by a user)
2>http://dx.com/p/rs232-to-usb-convertor-cable-2537?rt=1&p=2&m=3&r=2&k=1&t=1&s=48530&u=2537
(Prolific PL2303)

3>http://dx.com/p/usb-2-0-to-rs232-ii-serial-cable-adapter-converter-black-80cm-123347?rt=1&p=2&m=3&r=2&k=1&t=1&s=48530&u=123347
(FTDI based converters)

4>http://dx.com/p/usb-to-rs232-dongle-with-extension-cable-5859?rt=1&p=2&m=3&r=2&k=1&t=1&s=48530&u=5859
(Highest Numbers of reviews)

5>http://dx.com/p/bafo-usb-to-rs232-converter-5947?rt=1&p=2&m=3&r=2&k=1&t=1&s=123347&u=5947
(Best built Quality)

6>http://dx.com/p/usb-to-rs232-serial-port-adapter-transparent-green-24512
(Cheapest Price)

Where I have had the most problems with the serial converters is the driver (When I last checked all the supplied drivers was for Win XP), especially with Prolific PL2303.

I solved the problem by buying locally from a company that did supply working Win 7 drivers (Sandberg).

I cannot say which one of the adapter is best or works, that would require that I tested them.

I have ordered but not yet received my UT61E. I am reading the manual and doing other research prior to having my hands on the real meter.

The current measurement would indeed have a “funny” resistance. Looking at the UT61E’s schematic floating around on the web, it looks like they use shunts with dual diode in parallel depending on the range.

  • With the max (10A) range: 0.01ohms, two zener diodes back-to-back (~~~~) then in parallel.
  • With other ranges, shunts are: 9, 90, and 900 ohms, in series and 9/900 are in parallel with two diodes (one in reverse). The 900-90-9 are in series then series with a 2ohm where. Drops appears to be measured at left of 900 (900+90+9+2) or left of 9 (9+2) depending on setting

Since diodes are not linear devices, the resistance would depend on the voltage and current. So the resistance will not be a fixed value as in a regular resistor alone but a resistance that changes with I and V. I am no electronic expert. I am not sure exactly why the diodes are there. My best guess is the diodes are to limit the voltage drop across the shunt so it can switch to the higher range without having the lower ranges reading a voltage way too high for it. (Any expert out there? Enlighten me if my guess is wrong.)

If the schematic is right (found it on the web somewhere, doesn’t mean it is accurate): For the 10A range with 0.01ohm shunt, the burden voltage should not have much affect. A 10A current would drop a mere 0.1 volt drop. The 1.2% error at 10A would give a +120mA error, plus 50-count giving total error of + 170mA. Since the 50-count alone would represent 50mA, measuring anything below 1A would not be very good with a 50mA uncertainty minimum.

Next best is the mA range with 9ohm in parallel with 2 diodes then series with a 2ohms shunt. The diodes+shunt would manifest itself as somewhere below 11ohms - the diode makes the actual resistance current/volt dependent so how much below 11ohms is hard to say. Inserting that into a say direct drive LED with a 50ohm current limiter, the 11 ohms would introduce a 20% increase in current limit resistance - the 50ohms limiting resistor is now a 61ohms limiter. That current measurement will be rather useless.

Using a 0.1ohm shunt and the lowest voltage range for low current (220mv) would in theory give better measurement. The shunt will be able to measure up to 220mV which equates to 2.2amps. The accuracy at that range would be (according to the manual) 0.1% accuracy giving error at –0.22mV. The five count error makes total error–0.27mV or 2.7mA.

So, it looks like I need to find some 0.1ohm shunts.

I have also seen the schematic (now).

The diodes protect the unused range, i.e. is 90 and 900 ohms are in series, when using the 90 ohm resister the diodes are used to lead the current around the 900 ohm resistor. That is the reason for the "funny" resistance in ampere and it does not affect precision "only" burden voltage.

I hate flipping through the manual to look for the %accuracy/specs each time I change the dial on my meter. So, I cut and pasted the UT61E’s spec for all the selections into a single-page layout. Small prints, but I can see all the %accuracy for all the ranges on this little poster on my wall.

Thought other UT61E may find this useful. If you like it, feel free to use it.


I was going to upload the PDF file here, but I see there is no upload option. So here is a link to EEvBlog under UT61-E teardown photos where you can find the PDF file.

Link to all specs on a single page