Pros and Cons of owning a Fluke 8508A

[Inverted18650]

Hello all,

I am hoping you all will help me decide between the HP3458 and the FLuke 8508A. I have all the other Fluke 8500 series units and love them but this is certainly not a normal DMM or acquisition. The seller is willing to work with me and is giving me credit on trade for some gear I have here; so this would be a trade with cash deal.

If anyone owns both, I would love to hear the pros and cons and if you feel the premium for the 8508 is a worthwhile expense. I am working toward building my own small CAL LAB and this is probably a once or none deal for me. I know several of you have the 3458A and even the gold standards of that model and just would like some input before I make a "major" investment.

Much love and thank you all for sharing your ideas,

Chris

[ycui7]

I don't have any of these two equipment. However, I once have a need to measure 100MOhm resistor at 10 ppm accuracy.

I did some research and it turns out, the Fluke 8508A is the only equipment I found that capable of it, at least from the datasheet.

All other DMM whether 6.4 digit or 8.5 digit, all have 10Mohm shunt resistor when measuring 100MOhms and above resistance. This basically made them not suitable to get the accuracy to 10 ppm, mostly around 100 pm level or worse.

So, if you ever need measuring 100M or 1Gohm with high accuracy requirement, you should go with Fluke.

[ScoobyDoo]

Hello Inverted18650 / all - there are clear indications in Germany that Fluke will launch 2 new 8 1/2 DMM models in a few days from now (20th March 2019): 8558A and 8588A
There is also another  rumor that Keysight Germany will launch a successor for the 3458A ... - sources inside Keysight have confirmed this rumor too - it is like a Tom & Jerry game ...
Prices for the relaxed model 8558A will be in par with EU3458AX and for high end metrology model 8588A will be near 16k€-ish ...
I do not think it is the right moment to invest in existing models before you have seen specifications of new models and early birds have confirmed the stability of these instruments.
If you want to start a CAL Lab you will definitely need a high end Fluke 732B/732C/734 or Fluke/Wavetek 7000 10V source and 10kOhm standard references too.
There are hidden costs with these instruments - a good 10V DC metrology source is more stable than the best DMM available - maybe good to consider to maintain the 10V DC source ...
A logical effect would be that prices of used 8508A will drop also as Fluke will introduce their high end 8588A at lower cost than old 8508A
Check first if you want to maintain the calibration of your next DMM (certified) or do it yourself using traceable CAL standards and check which instrument supports in house calibration.
It is noticeable that all these (new) high end metrology multimeters have been developed in european labs ...

Herzliche Grüße/Meilleures salutations/Best regards
ScoobyDoo

[ap]

There are a few things to consider, besides just raw uncertainty data, e.g:
 
The 3458A is unbeaten in ADC linearity. So for precision voltage measurements where you use a voltage standard to transfer the voltage rather than relaying on the (then relatively poor) uncertainty of a 8.5 digit DMM, the 3458A is the clear choice.
The 8508A has tighter specs in some respects, is going higher in range (2 instead of 1.2), and has a specified transfer uncertainty for the ohms ranges, the 3458A has not (written specs are always important for a cal lab). The 1281 has similar properties btw, just compare specs.

So depending on what you do, you may need both. Option, if budget is limited, may be a used 3458A Opt 002 (or even the 2ppm one) and a 1281 at a total price below the 8508A new allone. Plus of course, if you buy a new one, you may wish to wait until the successors are available and beyond infant flaws.

[TiN]

F8508A does not go to 2.x ranges but limited at 1.9990 (so you cannot measure 20.0000V with 8508A  ). If you building own cal-lab I would consider getting F5700/5720A instead, it's far more useful for calibration work, as even if you have ten 8.5 DMMs, you still need high-stability source for calibration points at your DUT's.

Also without option 001 Fluke DMM is much less useful for resistance transfers. And like previous posts mention - get yourself ready for high annual calibration costs. In either case you need to have own multi-cell 10V standards, at least 10Kohm, 1 ohm standards, often 1Gohm standard, current shunt set, AC/DC transfer gear and list goes on and on... just to be able verifying 8508A or 3458A performance.

I spent quite some time with 8508A, and in my personal opinion it's not worth 2.5 times price difference to 3458A (based on new unit cost). So consider all aspects, unless you getting one for cheap.

[branadic]

Beside the technical specification differences between both meters here are some pros for the 8508A and thus cons for the 3458A that are also worth accounting for:

- F8508A has LCD displays, from my point of view the most reliable way of displaying data beside 7 segment led, instead of a VFD that will dim with time. Looks like the new meters will have color LCD displays and only time can tell how reliable they are.
- F8508A is the younger meter with modern design, while the HPAK3458A is more like a grandfather with replaced organs to keep it alive, but there was no brushstroke to give it a modern look and feel. Thus the user experience / user guidance of the 3458A is as old as the meter itself (~40 years?).

-branadic-

[TiN]

Yup, those are valid subjective points too. In typical callab I'd expect both controlled remotely. Thanks to programmable buttons, 3458A UI is actually easy to use, while 8508 have complex and clunky menu tree's, essentially result of reuse Datron/Wavetek 1281 UI logic.

Personally I prefer VFD with great contrast and view angles. 8508A's LCD is one of the worst for contrast. Taking photo for article with 8508A screen always need lot of fiddling with light/camera settings

8508A is also passively cooled, meaning that it will get hot and whatever equipment you locate on top/below it - will get unwanted thermal gradients.

So OP will hear as many biased comments with own merit to each. In the end it's best to have all meters, because when one of them drift away or fail , you could detect that and do some expensive repairs. 

[Dr. Frank]

This is just another case, where the T.O. should at first think about, and define/specify  the goals of what we wants to achieve.

Whenever I read ".. want to set up my own small CAL LAB.." and then directly ask for 8.5 DMM , that's the wrong approach, upfront.

Make yourself clear that, if you REALLY want to do electrical calibration, you at first need all kind of sources, i.e. for all different modes and ranges of a typical DMM. If you own either the 3458A, or the 8508A (or similar), these alone will not let you calibrate anything, because all the (stable) stimuli are missing...but both instruments are practically on par for most of the calibration tasks, as soon as you also have the appropriate calibrator, as TiN already stated.

If you just want to maintain and measure only the basic electrical units, (not to really calibrate anything), better get the 3458A, a 10V and a 10k standard, and you are done. ...because it's got the unique Artifact and ACAL features, which lets you precisely adjust all modes and ranges from only two sources.

And again, make yourself sure, which units / ranges you really need / want to measure, and to what accuracy.

And plan for cost for regular calibration of these instruments also.

Frank

PS: branadic, please do not exaggerate the vintage of the 3458A, it's from about 1988, so JUST 30 years old.

[branadic]

PS: branadic, please do not exaggerate the vintage of the 3458A, it's from about 1988, so JUST 30 years old.

Okay, still 31 year and thus an oldie

-branadic-

[Mr. Scram]

31 years is still a lot for a device on the cutting edge.

[Kosmic]

So, if you ever need measuring 100M or 1Gohm with high accuracy requirement, you should go with Fluke.

Usually electrometer are a lot better at measuring high resistance. I'm sure you could find one suiting your requirements. Might not be necessarily cheaper than the Fluke 8508a though.

[Inverted18650]

This is just another case, where the T.O. should at first think about, and define/specify  the goals of what we wants to achieve.

Whenever I read ".. want to set up my own small CAL LAB.." and then directly ask for 8.5 DMM , that's the wrong approach, upfront.

Make yourself clear that, if you REALLY want to do electrical calibration, you at first need all kind of sources, i.e. for all different modes and ranges of a typical DMM. If you own either the 3458A, or the 8508A (or similar), these alone will not let you calibrate anything, because all the (stable) stimuli are missing...but both instruments are practically on par for most of the calibration tasks, as soon as you also have the appropriate calibrator, as TiN already stated.

If you just want to maintain and measure only the basic electrical units, (not to really calibrate anything), better get the 3458A, a 10V and a 10k standard, and you are done. ...because it's got the unique Artifact and ACAL features, which lets you precisely adjust all modes and ranges from only two sources.

And again, make yourself sure, which units / ranges you really need / want to measure, and to what accuracy.

And plan for cost for regular calibration of these instruments also.

Frank

PS: branadic, please do not exaggerate the vintage of the 3458A, it's from about 1988, so JUST 30 years old.

Thank you Doc & everyone else for replying,

My goal is to be able to "calibrate" ( take a measurement against a known reference) anything 6.5 digits and below, mostly other DMM's. I have the standards ( IET/ Gen Rad: for resistance & capacitance) and a few other units for voltage and current. Right now I am working with the Fluke 5100B, 5200A, 3330B, 335D, 332D, HP 3245A, and my best meter is the Keithley DMM7510. ( I have a K2450 en route to replace my old K228's and I am shopping for a F5220A to get 20A output). I am sending the 5100B, 5220A, and HP3245A go through a fresh CAL from a recognized lab, once I get the K5220 AMP.

Again, I am just looking for pros and cons of owning the F8508A and it seems like my money is probably better spent getting 2 x HPAK3458A or one a few more standards, or save for the 5700 series as TiN suggested. I will also look at the 1281. (note: I arranged to buy a HPAK3458A from another forum user, but got myself banned and could not see nor reply to his messages and ended up loosing it... ). So I am trying to make the best of a bad situation and get more insight from the pros here.

Thank you all for the kind replies and I will continue to read and take the advice of the pros on the forum...MANY THANKS.

[quarks]

as said by others above, there are plenty pro and cons to 8508A.

As I have several 8.5 digit DMMs in my personal ranking, I prefer the Datron/Wavetek/Fluke design over 3458A.
Right now I would try to go for a 1281, mainly because the price of a 8508A is just way to high.
If you plan to buy new, you should wait until the new 8558/8588A will be out and hope that 8508A will get cheaper.

But when price/linearity/availabiltiy ist most important, a 3458A is very hard to beat.

[e61_phil]

If you just want to maintain and measure only the basic electrical units, (not to really calibrate anything), better get the 3458A, a 10V and a 10k standard, and you are done. ...because it's got the unique Artifact and ACAL features, which lets you precisely adjust all modes and ranges from only two sources.

I don't want to start a holy war again, but in my opinion the CAL 10, CAL 10k feature (lets take ACAL out of this discussion so far) isn't really that useful for a cal lab. You have to verify every range and function anyway and the CAL "feature" destroys all the history of the other ranges, if you have to readjust the instrument. Even if just a stupid high amp range is out of spec (we had such issues in the past).
In my opinion a meter which allows you to adjust every range independent from other ranges is much more useful in case of characterization.
If you want to have something to calibrate all ranges with only one 10V source I would search for an Fluke 752A.


Btw. The 8508A is a really great meter for Ohms and current. In this region it is much better than the 3458A. And it seems to be more stable in DCV, but if you have a stable 10V source anyway it isn't so important.

[Dr. Frank]

If you just want to maintain and measure only the basic electrical units, (not to really calibrate anything), better get the 3458A, a 10V and a 10k standard, and you are done. ...because it's got the unique Artifact and ACAL features, which lets you precisely adjust all modes and ranges from only two sources.

I don't want to start a holy war again, but in my opinion the CAL 10, CAL 10k feature (lets take ACAL out of this discussion so far) isn't really that useful for a cal lab. You have to verify every range and function anyway and the CAL "feature" destroys all the history of the other ranges, if you have to readjust the instrument. Even if just a stupid high amp range is out of spec (we had such issues in the past).
In my opinion a meter which allows you to adjust every range independent from other ranges is much more useful in case of characterization.
If you want to have something to calibrate all ranges with only one 10V source I would search for an Fluke 752A.


Btw. The 8508A is a really great meter for Ohms and current. In this region it is much better than the 3458A. And it seems to be more stable in DCV, but if you have a stable 10V source anyway it isn't so important.

Hey Philipp,
it's not a discussion (don't like 'holy war') about a certain instrument, but about the requirements of 'calibration'. 
In formal aspect, you're right, that even Artefact /ACAL instruments need a subsequent verification of this Artefact process. For practical (volt-nuts) use, anyhow, when only auto adjustment is required, nothing beats the 3458A or the 57x0 instruments.

I had to take up this discussion, as you're referring to the 752A, which itself is an autocal ratio instrument, just like the 3458A and the 57x0A, which also make ratio adjustments (beside other transfers), but the 752As 10:1 / 100:1 / HV adjustments are fully accepted to be a ratio calibration also, by design and error calculation only, with no further verification required, obviously.

Maybe you could explain, why this is treated differently than these other instruments? 

[e61_phil]

Regarding the "holy war" I just don't want to start a discussion about which instrument is best. If you have a Keithley meter which only saves your adjustments if you go through all the ranges and adjust them, you don't win anything over the artifact cal (it might be more close to the standards, but that's all).

As far as I know our local cal lab also calibrates their 752A frequently.
You're absolutely right that a Fluke 752A could also have a defect, but I would trust that method more than any ADC. And in most cases the uncertainty with a Fluke 752 is much lower. The resistors for higher voltages aren't that great in many meters (including 3458A).

If your meter has a good ADC you don't need this CAL feature. You could do the transfers by hand as a volt-nut.

[kj7e]

I'm still holding out for a DMM8510.

[ScoobyDoo]

Hello folks - here is already a short commercial of the 8588A



See you all soon on 20th of March at Fluke event in Kassel !

Herzliche Grüße/Meilleures salutations/Best regards

ScoobyDoo

[Magnificent Bastard]

Hello folks - here is already a short commercial of the 8588A



See you all soon on 20th of March at Fluke event in Kassel !

Herzliche Grüße/Meilleures salutations/Best regards

ScoobyDoo

I can't see the video in the USA (it says: "this video is unavailable").  I don't know if it was taken down, or if YouTube is limiting it's viewing to the EU.

[ScoobyDoo]

Hello Folks,





Herzliche Grüße/Meilleures salutations/Best regards

ScoobyDoo