looking for decent multimeters

[M. András]

hi, I'm looking for a multimeter for myself, home and for future possible work usage by myself at home

the benchtop what i was looking at are these:
fluke 8808a, 8845/46a
agilent 34405a, 34401a
any recommendation on keithley ones and others whit decent quality and can be expected to serve a lifetime? pricetag up to the fluke 8846a
handheld:
fluke 289, or the agilent's top one
if you own any of these what i listed i would be happy to read your opinion and experience  about these performance, issues etc. and I'm also intrested in that if any of these holds the accuracy more than 1 year without calibration

[alm]

Keithley 2000 is quite good and has specs very similar to the Agilent 34401A.

Any high quality meter is likely to be accurate for a long time, especially after some time (drift tends to decrease with time). A Fluke 8846A has a much higher 1 year accuracy (24ppm at 10V range) than a Fluke 8808A (150ppm), so you can also expect the accuracy after two years to be much better. It's fairly rare to see 2 year accuracy specified, the Agilent 3458A is one of the few current ones that comes to mind. If you want guaranteed specs, observe the recommended calibration intervals. If it's not that important, it's fairly safe that a good quality meter won't be far of its 1 year specs as long as it doesn't get damaged or modified. It's not uncommon to find old meters by the likes of HP that have not been calibrated for many years that are still in spec. Of course these have the advantage that drift usually stabilizes with age.

[grenert]

I've had the older Fluke 8840A, which was the precursor to the 45, which was the precursor to the 8808A.  It was a real beast, mechanically.  Beautifully made.  I noticed the 8808A is made in China now, so I'm not sure the build quality would be the same.  The 8846A is made in USA still. 

I have the 289 "handheld" meter.  It is a great meter, but it will probably not leave your benchtop because it is HUGE and HEAVY.  If that is OK, then I think you would like it.  Mine has stayed in spec (as far as I can tell) for a couple of years without calibration.  I think the best feature is event (changes in the reading, rather than recording at timed intervals) logging and being able to look at the result on the meter.  No need to mess with a PC.  However, note that you can't view a live graph of the data as its being logged.  One other drawback besides the size is the high resolution display.  I find a standard segmented LCD easier to read (and VFD even easier).  That said, with all of the features available on the 289, the hi-res screen allows for menu selection which is very intuitive.  Fluke leads (made by Pomona) are top quality.  Fluke charges an arm and a leg for their PC Flukeview Forms software and USB cable, so if you're interested in getting that, I would buy the package with the meter and the cable/software included.

If you are referring to the U1253A/B OLED meter as Agilent's top one, I don't have that meter but I do have the non-OLED version (U1252A)  which is otherwise essentially the same.  The user interface with the small, strangely placed buttons is weird to me, compared to Fluke's layout.  I think the fast update speed of the display is supposed to be a feature, but I find the rapidly-changing numbers more of a distraction.  It seems to be very accurate, but I haven't had mine long enough to say anything about long-term stability.  The test leads/clips that it comes with are decent quality, but are not in the same league as Pomona.  Agilent gives their PC software for the meter away free.  I assume you've seen Dave's video review of this meter.

Regarding the Agilent bench meters, I have no firsthand experience with any of them, but I have been looking at them, too.  If you are going to spend the money for a 34401A, the 34410A is not much more money for a much more modern meter.  It has USB and a dual display, as well as advanced logging features.  It also has capacitance and temperature, which the 01A doesn't have.  I don't see a good reason to get the 34401A unless you need to be consistent with other equipment you already have.

[EEVblog]

The (rebadged I think) Keithley bench meter has some very bad reports on various forums, beware.

Dave.

[robrenz]

I looked at all the meters you listed before I bought a new 8846A.  For my purposes of bench work only (not automated system measurement) the  8846A has the most bang for the buck. It has wider ranges on many items and more unique features and has USB connectivity. There is no comparison between the Fluke and agilent displays. The Fluke display is much more readable and informative and the trend plot feature is really useful.  I really like the meter it is even better than what I expected from reading the manual.  The operation of the meter is much more intuitive than I expected.
Just my opinion for what its worth, I am sure you will get others.

[alm]

The (rebadged I think) Keithley bench meter has some very bad reports on various forums, beware.

You're confusing the Keithley 2000 series with the Keithley 2100. The latter was a rebadged (Array?) design with a fair number of firmware issues, the 2000 series was introduced before the 2100 and I haven't seen any negative reports, please post a link if you have. This is why I specifically mentioned the model 2000. The 2001, 2002 and 2010 are also nice, but much more expensive.

I think the main point in favor of the Agilent 34401A is that it has been the industry standard 6.5 digit bench meter for ages, while the Agilent 34410A and the Fluke 8846A are recent designs. This is why the Fluke 8846A can emulate the Agilent 34401A remote protocol. In the case of Fluke, they re-entered the market after they were out of this market segment for ten years or so. Nope, the Fluke 45 or rebadging Agilent equipment with a single resistor changed doesn't count. Another point in favor of the 34401A is that the service manual contains a complete schematic, something which is not the case for newer designs as far as I know. Agilent is pushing quite hard to replace the 34401A by the 34410/11A, so my guess is that the only reason why they still produce is because it's part of tons of procedures, and companies may be hesitant to rewrite/recertify those procedures and retrain staff.

The only substantial negative I've heard about the Fluke 8845/8846A is a complaint about the split banana jacks not being very durable. One spec that Fluke doesn't do very good on is the reading rate. Max. 1000/second at 4.5 digits. This matches the old Agilent 34401A, is less than the 2000/second by the Keithley 2000 and is much less than the 10,000/s at 5.5 digit by the Agilent 34410A. This spec is mainly relevant for sampling or system use. Or characterizing parts, where you may have to cycle through all 2ⁿ codes of a DAC. The Fluke 8846A is also more expensive than the Agilent 34401A, 34410A or Keithley 2000.

I'm not trying to argue in favor or against any of these meters, just trying to give some more details.

[M. András]

split banana jacks are used mainly for their special 2 lead 4wire resistance probe? but it will last for years if you dont plug/unplug them 30 times a day i guess (if not my choice would be that i make a very short extension cord with high quality banana plug/wire socket and wire, not so nice but it would be still better then that 750usd repair cost what i read here somewhere)

i was checking farnell's prices when i made this list, unfortunately the only distributor in the country for keithley, is strictly sells only to companies.
anyway so from keithley the 2000 series would be good choice too.
that 8 hours battery life on the 1253 series is a joke, if it would be the only meter what i have and i only use it on the bench thats damn sure i would wire it to some lithium cells or a power supply, i will have time to think and talk about them to know more from each of these instruments before i will have the money for it.(i hope any of the handheld ones will be bought in 2-3 months)
anyone have any information what are the changes between the 289 and the 289 eur model? the 289 eur is cheaper but internally whats changed? if anyone have an info from this

[grenert]

One thing I've noticed looking at current bench meters is the usage of the secondary display.  The Fluke 8808A and 8846A, and the Agilent 34405A appear to be able to show current and voltage simultaneously, while the higher-end Agilent 34410A cannot.  That is really a shame, and I'm surprised that the lower end model has a feature lacking on a more expensive one.  I was pretty well set on the 34410A, but now I am more seriously considering the Flukes.

[robrenz]

2nd display readings on the  8846A

MAIN    -  SECONDARY
DC volts - AC riding on the DC
AC volts -  DC offset of AC or Frequency of AC
Frequency - Period
Period - frequency
DC current - AC current
DC current - DC volts
AC current - DC current AC current is riding on
Temperature - Ohms

The 2x4 wire kelvin probes are pretty nice.  You can probe smd resistors right thru the carrier tape front or back. I asked fluke about them and they said the source and sense lines were kept separate as close to the probe tip as physicaly possible.  The measurements agree very closely with my regular Kelvin clip leads.

[grenert]

Thanks, Rob.  Here is all that the 34410A can do:

Primary Function  -  Second Display Function
DC V, DC I, AC V, AC I  -  Peak–to–Peak value
Freq  -  AC Voltage
Temp  -  Raw Resistance

It's so limited that I would not really even consider the secondary display a feature.   

[robrenz]

I know I am sounding like a Fanboy but here are some of the things that made me choose the 8846A over the 34410 or 11

8846A      33411
10 AMPS                 3 AMPS     
1MHz                      300KHz
100 MILI FARAD      10 MICRO FARAD
1000 AC VOLTS       750 AC VOLTS
10 OHM                  100 OHM

8846A ONLY
Diode test selectable 5V or 10V  1 mA or 100 microA
Continuity selectable threshold 1 Ohm to 1000 Ohm 300 samples/sec with audible
MIN MAX AVG. SDEV # OF SAMPLES all displayed simultaneously
Manual offset input
MX+B of input
Trend Plot (Realy nice)
Histogram display
DCV RATIO of 2 voltages with single common uses three leads.
USB PORT extended memory stick stores up to 50000 readings + 10000 readings in internal memory.

The 33411 measures up to 50000 readings per second, but at .001 power line cycle.  the Fluke measures maximum 995 readings per second at .02 NPLC  You have to decide if readings at .001 NPLC are accurate enough.

[Kiriakos-GR]

Those Picometer (brand) looks well priced for what they do , give them a look.   

[ErikTheNorwegian]

I have a 8846A, and i must say, worth every penny/krone..
I love this meter, and its just everything i can want i a bench meter.
Only thing i don't like is that it starts in DC volt and flashes overload on and off, due to sensitivity of the DC mV when the measuring cables are connected...

[alm]

I know I am sounding like a Fanboy but here are some of the things that made me choose the 8846A over the 34410 or 11

Not sure why you would compare the 8846 to the 34411, the 34411 completely blows the Fluke away in reading rate, and is equal to the 34410 in all other regards if I remember correctly.

Continuity selectable threshold 1 Ohm to 1000 Ohm 300 samples/sec with audible
Manual offset input
DCV RATIO of 2 voltages with single common uses three leads.

You can also set the null value on the Agilent 34401/10A. The DCV ratio and continuity are interesting omissions from the 34410A, since the 34401A does support them. The Fluke continuity spec was probably directly copied from the Agilent 34401A, which has exactly the same spec. The Keithley 2000 also has variable continuity threshold, multiple diode test ranges up to 10V, mX+b and null math functions. No fancy display modes or USB connectivity, however.

The 33411 measures up to 50000 readings per second, but at .001 power line cycle.  the Fluke measures maximum 995 readings per second at .02 NPLC  You have to decide if readings at .001 NPLC are accurate enough.

It's hard to measure 50,000 reading per second at 0.02 NPLC, unless your mains frequency is 1 kHz or so. I believe the noise performance is quite good, but NMRR will off course be reduced. The Agilent 34410/11A will do 3000 readings per second at 0.02 NPLC, and various other speeds so you can trade off noise performance for speed. If this is an important spec for you, the Agilent 3441xA series is the only option (apart from the expensive Agilent 3458A) as far as I know.

Note: the only reason why I only mention negatives of the 8846A is that you only mention positives, not because I think the Agilent 34410A is a better meter. Based on specs (I haven't used the Fluke 8846A) the Fluke 8846A appears superior in almost all aspects, but at a higher price.

Only thing i don't like is that it starts in DC volt and flashes overload on and off, due to sensitivity of the DC mV when the measuring cables are connected...

That's a common issue with bench meters. It doesn't take a lot of current to induce a few volts at > 10 Gohm input impedance.

[robrenz]

To show I might have a small bit of objectivity,

The  33411 is slightly more accurate on frequency and capacitance, slightly less accurate on DC volts
33411 has offset compensated ohms, very nice.  You can accomplish the same thing with current reversal on the 8846A but a lot more work.

8846A DC current burden voltage
100 micro A <0.015 V
1 mA 1<0.15 V
10 mA  <0.025 V
100 mA <0.25 V
400 mA  <0.50 V
1 A <0.05 V
3 A <0.15 V
10 A <0.5 V

33411 DC current burden voltage
100.0000 micro A <0.03 V
1.000000 mA <0.3 V
10.00000 mA <0.03 V
100.0000 mA <0.3 V
1.000000 A <0.80 V
3.00000 A <2.0 V

[robrenz]

Not sure why you would compare the 8846 to the 34411, the 34411 completely blows the Fluke away in reading rate, and is equal to the 34410 in all other regards if I remember correctly.

I think the reading rate is irrelevant for an actual Bench meter (not system use).  If you need more than 995 readings per second the 33411 is the meter for you. but I cant decipher (I don't have the technical chops to figure it out) from the manual what the accuracy would be at that speed but I can see there is no normal mode rejection until you get to 1 NLPC and it looks like the accuracy really goes to pot at 50000.
The reading rate and offset ohms are the only pluses I see for the 34411

[alm]

My point was: either reading rates are important to you and you get the Agilent 34411A, or they're not and you get the Agilent 34410A or one of the other options. I don't see an application where either the Fluke 8846A or the Agiletn 34411A would work, but not the Agilent 34410A.

Of course the NMRR is zero at <1 NPLC, how do you reject periodic signals at less than a single period? Does any of the competitors do better? The solution is simple: don't have AC superimposed on your DC signals.

Accuracy is quite good for 4.5 digits, about 0.01% of range extra uncertainty on most ranges. Better than the competition at similar reading rates, I think.

[robrenz]

Only thing i don't like is that it starts in DC volt and flashes overload on and off, due to sensitivity of the DC mV when the measuring cables are connected...

You can store your favorite meter configuration as the power up configuration.
 
Storing the Power-up Configuration
To store the Meter’s present configuration as a power-up configuration:
1. Press (memory) button
2. Press the STORE CONFIG softkey.
3. Press the STORE POWER-UP softkey.
The Meter configuration stored as a power-up configuration will be set whenever the
Meter is powered on.

You can also store 5 different configurations in internal memory and 99 configurations in external if you have a memory stick in the usb port

[M. András]

dual display sounds good and i listed it in my paper in front of me the 8846a looks like have the most of the function for the 2nd display, i didnt check it can store 10k readings in its memory or i missed it in the manual/spec sheet, and it makes a graph from those or it would display them as invidual readings?

[robrenz]

dual display sounds good and i listed it in my paper in front of me the 8846a looks like have the most of the function for the 2nd display, i didnt check it can store 10k readings in its memory or i missed it in the manual/spec sheet, and it makes a graph from those or it would display them as invidual readings?

details of trend plot
http://www.myflukestore.com/crm_uploads/fluke_trendplot_-_data_logging_on_the_fluke_884x_with_trendplot_application_note.pdf

[M. András]

dual display sounds good and i listed it in my paper in front of me the 8846a looks like have the most of the function for the 2nd display, i didnt check it can store 10k readings in its memory or i missed it in the manual/spec sheet, and it makes a graph from those or it would display them as invidual readings?

details of trend plot
http://www.myflukestore.com/crm_uploads/fluke_trendplot_-_data_logging_on_the_fluke_884x_with_trendplot_application_note.pdf

thank you 

[robrenz]

This also may be helpfull
http://support.fluke.com/calibration-sales/Download/Asset/2547797_6203_ENG_B_W.PDF

[M. András]

This also may be helpfull
http://support.fluke.com/calibration-sales/Download/Asset/2547797_6203_ENG_B_W.PDF

i read this before but thanks

[M. András]

anyone have any info from the fluke differences between the fluke 289 and the 289 eur model, what they changed etc.

[IanB]

anyone have any info from the fluke differences between the fluke 289 and the 289 eur model, what they changed etc.

I think they changed the probes for new ones with more shielding over the metal parts to satisfy newer safety standards.

Eventually the safety standards will require the probes to be entirely enclosed in plastic so that no metal parts are exposed at all. Then they will be completely safe