Author Topic: Fluke 8846a teardown  (Read 14969 times)

0 Members and 1 Guest are viewing this topic.

Offline WartexTopic starter

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ca
    • http://headsplosive.com
Fluke 8846a teardown
« on: November 13, 2011, 04:52:06 am »
http://headsplosive.com/2011/11/fluke-8846a-6-5-digit-multimeter-teardown/

Before anyone starts bitching I didn't embed video directly into the forum - I pay a lot of money for my gear and make about a quarter of what Dave makes, I need to recover the costs thru traffic to my site. There is many more reviews and teardowns coming of expensive gear such as Agilent bench DMM, Fluke thermal imagers, spectrum analyzers and so on.

I'm new to electronics so don't laugh if I screw up identifying the components.  :P
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Fluke 8846a teardown
« Reply #1 on: November 13, 2011, 06:06:58 am »
I got mine new from tequipment.net for US$1282.50 you have to ask for a quote to get that pricing.  Their site shows US$1425.00 normally. I also got $200.00 worth of free fluke tools ( ac votage detector, infrared temp gun, and fluke tote bag) for spending that much  during a promotion.  Awesome meter!

Offline Richard W.

  • Regular Contributor
  • *
  • Posts: 149
  • Country: de
Re: Fluke 8846a teardown
« Reply #2 on: November 13, 2011, 11:33:30 am »
The LM399 is the buried zener voltage reference with a heating.  It's responsible for the stability of the meter.

Richard
 

Offline phil_jp1

  • Regular Contributor
  • *
  • Posts: 103
  • Country: ua
Re: Fluke 8846a teardown
« Reply #3 on: November 13, 2011, 02:27:20 pm »
I didn't expect Fluke to make those last-minute-repairs (if, of course it is their job). But, I guess, they've decided that those few repairs would be cheaper than manufacture another batch of PCBs. Or maybe they found out about that problem after PCBs were all assembled. Who knows..

When you said, that there's nothing interesting inside and you didn't know why the price is so high, what were you expecting? Micro-circus-with-humpsters-running-in-the-wheels-generating-energy-for-those-precision-amplifiers?  ;D

Only those laser-trimmed precision resistor networks cost a fortune! Plus bunch of precision op-amps, which are pretty expensive; LM399 precision voltage reference; FPGAs, precision ADCs, and other stuff... it all expensive!

Plus you paying not only for parts, but for software, engineering expenses, etc. Not just because it's Fluke :) Of course our Chinese friends will make the same meter with the same characteristics, build quality and well written software for only third of a price!  ;)


Nice construction, as expected from Fluke. I like how they designed input selection switch directly on a PCB. With proper gold-plated polygons and contacts it should have pretty low resistance.
Also I noticed that there's a bunch of devices in TO220 packages, which just soldered in but not secured in any other way. I don't know, maybe they designed this meter only as a lab equipment and not for some production environment, where it can be subjected to vibration? (well, looks like it)
http://JumperOne.com - Electronic projects, tutorials, hacks, etc.
 

alm

  • Guest
Re: Fluke 8846a teardown
« Reply #4 on: November 13, 2011, 03:53:05 pm »
Before anyone starts bitching I didn't embed video directly into the forum - I pay a lot of money for my gear and make about a quarter of what Dave makes, I need to recover the costs thru traffic to my site. There is many more reviews and teardowns coming of expensive gear such as Agilent bench DMM, Fluke thermal imagers, spectrum analyzers and so on.
Thanks for doing the tear down. I don't see why linking would be a problem. There's nothing wrong with linking to your own website if you provide relevant, original content, that's what the WWW is about.

I didn't expect Fluke to make those last-minute-repairs (if, of course it is their job). But, I guess, they've decided that those few repairs would be cheaper than manufacture another batch of PCBs. Or maybe they found out about that problem after PCBs were all assembled. Who knows..
I'm not impressed either.

When you said, that there's nothing interesting inside and you didn't know why the price is so high, what were you expecting? Micro-circus-with-humpsters-running-in-the-wheels-generating-energy-for-those-precision-amplifiers?  ;D
Philips actually built a meter which used a motor and gears to switch ranges, just replace the motor with a hamster and you're there ;).

Only those laser-trimmed precision resistor networks cost a fortune! Plus bunch of precision op-amps, which are pretty expensive; LM399 precision voltage reference; FPGAs, precision ADCs, and other stuff... it all expensive!
Don't forget that some of the components, at least the LM399, are likely selected and aged.

Plus you paying not only for parts, but for software, engineering expenses, etc. Not just because it's Fluke :) Of course our Chinese friends will make the same meter with the same characteristics, build quality and well written software for only third of a price!  ;)
Really? Last time I checked the Chinese competitors like Rigol had inferior parts and were not that much cheaper, a few hundred bucks at most.

Also I noticed that there's a bunch of devices in TO220 packages, which just soldered in but not secured in any other way. I don't know, maybe they designed this meter only as a lab equipment and not for some production environment, where it can be subjected to vibration? (well, looks like it)
Lab equipment tends to be build to lower standards than handhelds as far as shocks and environmental conditions are concerned, since they are less likely to get dropped or banged around.

I was surprised by the CR2032 backup battery, I thought these were obsolete now we have non-volatile RAM like EEPROM.
 

Offline phil_jp1

  • Regular Contributor
  • *
  • Posts: 103
  • Country: ua
Re: Fluke 8846a teardown
« Reply #5 on: November 13, 2011, 06:30:12 pm »
Plus you paying not only for parts, but for software, engineering expenses, etc. Not just because it's Fluke :) Of course our Chinese friends will make the same meter with the same characteristics, build quality and well written software for only third of a price!  ;)
Really? Last time I checked the Chinese competitors like Rigol had inferior parts and were not that much cheaper, a few hundred bucks at most.
I was a bit sarcastic there  :)

I was surprised by the CR2032 backup battery, I thought these were obsolete now we have non-volatile RAM like EEPROM.

That battery is for Time and Date. Fluke 8846a has a clock.
http://JumperOne.com - Electronic projects, tutorials, hacks, etc.
 

Offline WartexTopic starter

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ca
    • http://headsplosive.com
Re: Fluke 8846a teardown
« Reply #6 on: November 13, 2011, 09:02:52 pm »
When you said, that there's nothing interesting inside and you didn't know why the price is so high, what were you expecting? Micro-circus-with-humpsters-running-in-the-wheels-generating-energy-for-those-precision-amplifiers?  ;D

Only those laser-trimmed precision resistor networks cost a fortune! Plus bunch of precision op-amps, which are pretty expensive; LM399 precision voltage reference; FPGAs, precision ADCs, and other stuff... it all expensive!

Plus you paying not only for parts, but for software, engineering expenses, etc. Not just because it's Fluke :) Of course our Chinese friends will make the same meter with the same characteristics, build quality and well written software for only third of a price!  ;)


My motherboard, ASUS P6T Deluxe v2 is made in china and has about 10x the complexity and component cost of the Fluke parts, and cost only $300. Most traces are transmission lines, and the whole thing runs at 2.5 GHz (pcie clock speed). The board has 7+ layers. So please, don't impress me with "oooo FPGA are expensive"! I just looked on Digikey, these FPGAs are $11 a piece.
My northbridge/southbridge is expensive too (to the tune of $60), and years of R&D that go into these devices. My whole PC cost less than this DMM, and has 1000x the functionality.

This DMM is overpriced by about 50%. There is $200 in parts MAX.

As for Chinese - I will tell you from first hand experience - the build quality is what american order makers are paying for. Chinese can make high end PC motherboards, and they can make cheap DMMs with shoddy soldering - it all depends on how much money you pay to the factory. They build crap because you are willing to buy it. I haven't seen a single piece of decent test gear in your videos. You knock China and then turn around and support low end manufacturers by buying their products, tsk tsk!
 

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 802
  • Country: au
Re: Fluke 8846a teardown
« Reply #7 on: November 13, 2011, 10:15:57 pm »
hello,
It's expensive yes but not every household has one. So theres not much demand as oppose to computer parts. These are low volume, low demand items. Hence you see there are only a few struggling manufacturers out there. You are buying many years worth of studies and research. The accuracy for a DIY person means there's little need to upgrade in the future. Count yourself lucky they are still around during this dark financial times, I'm sure its worth a lot more than what you paid in the long run.
 

Offline phil_jp1

  • Regular Contributor
  • *
  • Posts: 103
  • Country: ua
Re: Fluke 8846a teardown
« Reply #8 on: November 14, 2011, 03:02:21 am »
When you said, that there's nothing interesting inside and you didn't know why the price is so high, what were you expecting? Micro-circus-with-humpsters-running-in-the-wheels-generating-energy-for-those-precision-amplifiers?  ;D

Only those laser-trimmed precision resistor networks cost a fortune! Plus bunch of precision op-amps, which are pretty expensive; LM399 precision voltage reference; FPGAs, precision ADCs, and other stuff... it all expensive!

Plus you paying not only for parts, but for software, engineering expenses, etc. Not just because it's Fluke :) Of course our Chinese friends will make the same meter with the same characteristics, build quality and well written software for only third of a price!  ;)


My motherboard, ASUS P6T Deluxe v2 is made in china and has about 10x the complexity and component cost of the Fluke parts, and cost only $300. Most traces are transmission lines, and the whole thing runs at 2.5 GHz (pcie clock speed). The board has 7+ layers. So please, don't impress me with "oooo FPGA are expensive"! I just looked on Digikey, these FPGAs are $11 a piece.
My northbridge/southbridge is expensive too (to the tune of $60), and years of R&D that go into these devices. My whole PC cost less than this DMM, and has 1000x the functionality.

This DMM is overpriced by about 50%. There is $200 in parts MAX.

As for Chinese - I will tell you from first hand experience - the build quality is what american order makers are paying for. Chinese can make high end PC motherboards, and they can make cheap DMMs with shoddy soldering - it all depends on how much money you pay to the factory. They build crap because you are willing to buy it. I haven't seen a single piece of decent test gear in your videos. You knock China and then turn around and support low end manufacturers by buying their products, tsk tsk!

hello,
It's expensive yes but not every household has one. So theres not much demand as oppose to computer parts. These are low volume, low demand items. Hence you see there are only a few struggling manufacturers out there. You are buying many years worth of studies and research. The accuracy for a DIY person means there's little need to upgrade in the future. Count yourself lucky they are still around during this dark financial times, I'm sure its worth a lot more than what you paid in the long run.

Nukie right about that! The volume is completely different. And amount of money and human resources spent on motherboard (name a computer part) R&D is many times higher than on that multimeter.
P.S. That's right I don't have, pretty much any, expensive gear at home. I don't really need it there. It's not about what bike you ride, it's about how.

I just looked on Digikey, these FPGAs are $11 a piece.

I guess you used the wrong DigiKey. Those FPGAs are $41 a piece.

This DMM is overpriced by about 50%. There is $200 in parts MAX.

Well, if only those FPGAs alone cost about $80... then what about precision laser-trimmed resistor networks, ADCs and other parts? (don't forget the display, case, all the connectors, etc.)
And, in fact, a lot of those parts are custom made for fluke (in comparatively small volumes).
http://JumperOne.com - Electronic projects, tutorials, hacks, etc.
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Fluke 8846a teardown
« Reply #9 on: November 14, 2011, 03:26:29 am »
Part of that price is excellent timely support and being actually made in the USA. Mine was built on order by Fluke in their Everett Washington plant.  They do their own circuit boards and sheet metal right there and they do their own volt reference aging there also.  I called the plant for technical support twice before purchasing and twice so far after.  Each time in less than two minutes I talked with a very qualified technician that was able to answer my all my questions thoroughly.  I cant speak for other brands but I think that kind of support is uncommon today.  The price is very competitive with other meters in this class.

Offline WartexTopic starter

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ca
    • http://headsplosive.com
Re: Fluke 8846a teardown
« Reply #10 on: November 14, 2011, 04:39:04 am »
Quote
Nukie right about that! The volume is completely different. And amount of money and human resources spent on motherboard (name a computer part) R&D is many times higher than on that multimeter.

So, you just contradicted yourself. That's correct, motherboard require much higher R&D and tolerances than DMMs of 8846a class, which makes them more expensive to produce.  Also, that particular motherboard is FAR, FAR from common household board. So NO, the volume analogy is not really good here. It's a premium quality board made in limited quantities for gaming and overclocking enthusiasts, and there are literally hundreds of cheaper alternatives from other manufacturers, so it's not like it's going to have large sale volumes. It is not run of the mill Foxconn mobo. And trust me, there is enough labs around the world that buy those bench meters. The serial number span is in hundreds of thousands.

I think it's dumb to deny that there is a hefty markup on this DMM and you can have an equivalent from Agilent or some other brand for less. Even at $40 for FPGA (which they are not if you buy large lots) this still does not grant $1200 tag. You are paying for the name, and you are paying because there is not as much competition as in motherboard market.


Quote
P.S. That's right I don't have, pretty much any, expensive gear at home. I don't really need it there. It's not about what bike you ride, it's about how.
Is that what you tell your girlfriend?

Yes, making a thermostat with a simple PIC doesn't require precision DMM, I get it, but don't embarass yourself with bike analogies.  ;D
 

Offline Zad

  • Super Contributor
  • ***
  • Posts: 1013
  • Country: gb
    • Digital Wizardry, Analogue Alchemy, Software Sorcery
Re: Fluke 8846a teardown
« Reply #11 on: November 14, 2011, 05:42:49 am »
Are you saying the cost of an item is what all the individual components are priced at, plus someone assembling them? Because that is around 10% of the retail price of an item. If I have to design an item which retails at (say) £100, they will be sold by the manufacturer at approximately £25. In turn, it will cost the company around £10 in components and assembly. All the rest is development (quite a small percentage), marketing, administration, management, storage, support, power, light, heat, legal services... You get the idea.

There is an old saying "Knowing where to tap". If you haven't heard it before, please Google it.

Anyway, thanks for the teardown. I have a Tektronix DMM4020 5.5 digit bench meter in for test and evaluation (this is a rebadged Fluke 8808a) and am reluctant to remove the covers and break the cal seals. From what I can see through the side vents, they share a common architecture. It also uses an LM399H precision reference and trimmed precision resistors. Next time I have a good look through the vents, I will keep an eye out for any possible FPGAs. I cannot see any patch wires, and from memory the PCB is revision 10, so it has probably been around in one form or another for some time.




Offline WartexTopic starter

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ca
    • http://headsplosive.com
Re: Fluke 8846a teardown
« Reply #12 on: November 14, 2011, 06:42:44 am »
Are you saying the cost of an item is what all the individual components are priced at, plus someone assembling them?

No, I'm not saying that. Typical price for something like this is 2x or 3x the cost of manufacturing. Distribution sells this for $1200, meaning at a reasonable markup, they buy it for under 1k. That tells me that the cost of manufacturing (including R&D and all that other fluff) should be $300-400. But it's not. It's a lot less. The whole R&D argument is very, very dumb these days because there is not a team of 50 people poring over the PCB layout alone. These things are made by very small teams and very cheaply, with tons of automation software and prebuilt blocks. For example, Iphone 4 component cost is $70 (confirmed, look it up). Manufacturing, assembly and shipping is about $40. Let's say OS is $100 (this is VERY generous considering Windows 7 is $100 and is much more complex than iOS). The phone sells for over $600. Don't tell me there is $400 worth of R&D in the phone, please.

My point is, there are devices which are much more complex and cost a lot less. Fluke is pricing these for the market, not what they are really worth. Also I'm getting tired of fanboyism here. Yes, I like Fluke too and own a lot of their gear and love it, but let's face it their stuff is overpriced, just like Gossen and Agilent's software unlocks for new scopes.
 

alm

  • Guest
Re: Fluke 8846a teardown
« Reply #13 on: November 14, 2011, 11:38:57 pm »
Also, that particular motherboard is FAR, FAR from common household board. So NO, the volume analogy is not really good here. It's a premium quality board made in limited quantities for gaming and overclocking enthusiasts, and there are literally hundreds of cheaper alternatives from other manufacturers, so it's not like it's going to have large sale volumes. It is not run of the mill Foxconn mobo. And trust me, there is enough labs around the world that buy those bench meters. The serial number span is in hundreds of thousands.
A quick search indicates that discrete PCIe graphics cards (average price $300), which serve a similar market, have a sales volume of about 3M per year. Sure, only a fraction of them will buy an expensive mainboard, but it's still in the same order of magnitude as the total volume Fluke shipped since its introduction in 2006 according to your information. Many parts on a mainboard are common with cheaper models that sell in much larger volumes. The only models that can share parts with the Fluke 8846A are the 8845A and maybe the 8808A, which don't sell in huge numbers either.

You're probably a digital designer who looks down on his analog colleagues? A bus that shuffles 2.5Gb/s is much harder than a ~24-bit ADC that can just do a few thousand conversions per second, and not even 24-bit at that rate? Do you also consider a 24-bit/192kHz sound card superior? It can generate more bits in the same time span.

I think it's dumb to deny that there is a hefty markup on this DMM and you can have an equivalent from Agilent or some other brand for less.
Then buy the Agilent. Agilent is a better known brand than Fluke in the bench meter market. Fluke was essentially absent for a long time between the 8840/8842 (late eighties to early nineties?) and the more recent 8845/8846 series. The only thing they had was the 4.5 digit Fluke 45 and a rebranded Agilent reference DMM with a single resistor added. The Agilent 34401A is still the industry standard, although it's being replaced by the 34410A/34411A. The Agilent 3458A is the current reference standard. The Agilent 3456A still sees some active use, although I guess only if calibration is not required since that's gotta be expensive with all those trimmers. I'm not sure why Fluke would have a markup over Agilent, unless they offer betters specs. Agilent also has the advantage of offering other bench equipment like scopes, power supplies, function generators and almost anything related to electronic measurement.

No, I'm not saying that. Typical price for something like this is 2x or 3x the cost of manufacturing. Distribution sells this for $1200, meaning at a reasonable markup, they buy it for under 1k. That tells me that the cost of manufacturing (including R&D and all that other fluff) should be $300-400. But it's not. It's a lot less.
So were are all the competitors jumping in this hole and selling a similar meter for $500, which should be reasonable based on your $200 estimate? You can get cheap crap, similar price to real brands but inferior quality or similar price and close in quality (I've heard decent things about Picotest, but I wouldn't count on reliability or long term stability). Are companies like Rigol protecting Fluke's and Agilent's profit margins? They didn't appear to have any problem with undercutting them with scopes.
 

Offline WartexTopic starter

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ca
    • http://headsplosive.com
Re: Fluke 8846a teardown
« Reply #14 on: November 15, 2011, 01:18:29 am »
So were are all the competitors jumping in this hole and selling a similar meter for $500, which should be reasonable based on your $200 estimate?

Oh, they are there, trust me. Pretty much everything BK Precision sells is a rebranded tool made by someone else and sold for half the price, and BK is a brand name. LCR meters? Tonghui. Function Generators? Telulex. Power Supplies? Mastech/Xantrex. Spectrum analyzers? Micronix. Programmable loads? Just look for "bk dc load" on ebay and you will see unbranded ones that BK sells for 2x-3x the price. I took pretty much all of these apart and they are identical. The assembly quality is exact same (not bad).

Do you think Agilent or Fluke designs all their stuff 100% in house? Come on! Tonghui and Rigol alone are rebadged by many brands.

So yes, they are out there if you look for them, and they are coming out of the woodwork, like RIGOL did. My 20 MHz telulex func gen cost me $150 used. Equivalent used BK precision is 3x that. They are identical inside (including software). The are there, just look harder, and you can score some quality gear for less.
 

alm

  • Guest
Re: Fluke 8846a teardown
« Reply #15 on: November 15, 2011, 01:58:16 am »
OK, so where can I buy a $500 6.5 digit DMM similar to a Fluke 8846? That basically was your claim, wasn't it?

I'm not so sure if I'd consider BK precision a brand name, certainly not in the same league as Keithley or Tek. Yes, the big brand names sometimes rebrand equipment from the cheaper brands, although usually not with good results. The Agilent-branded Rigol scopes had bad documentation and support compared to the Agilent designs. The Keithley 2100 was designed and build by Array and sucked with buggy firmware which Keithley was unable to fix. Tek sold some cheap test equipment (GW-Instek?) in the nineties. No service documentation was available (not even for their own repair department) and quality was mediocre.

Which of the current Fluke or Agilent bench meters are rebranded Tonghui/Rigol/etc equipment? I'm not aware of any current equipment. The Agilent Rigol scopes were discontinued. Some of the old Escort designs may share a common heritage with other brands, but none of their flagship products as far as I know.
 

Offline Kiriakos-GR

  • Super Contributor
  • ***
  • !
  • Posts: 3525
  • Country: gr
  • User is banned.
    • Honda AX-1 rebuild
Re: Fluke 8846a teardown
« Reply #16 on: November 15, 2011, 02:08:03 am »
You're probably a digital designer who looks down on his analog colleagues? A bus that shuffles 2.5Gb/s is much harder than a ~24-bit ADC that can just do a few thousand conversions per second, and not even 24-bit at that rate? Do you also consider a 24-bit/192kHz sound card superior? It can generate more bits in the same time span.

It is superior, like more detail at high and low frequencies, that even boost the stereo separation a bit more.
The problem is that very few people have such sensitive hearing, so to spot the difference.
Another plus of the 24-bit/192kHz sound card, is that when your record at this sampling rate the latency gets lower.
I do write music, with my midi keyboard controller and the Cubase software, occasionally.   
 
 

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 802
  • Country: au
Re: Fluke 8846a teardown
« Reply #17 on: November 15, 2011, 01:57:24 pm »
I there are only a few current operating 6 digit multimeter manufacturer out there. Like alm mention, Picotest is the most rebranded unit. Advantest, Fluke/Tek and Agilent. There rest is just crap even with high accuracy spec, long term drift is not documented due to lack of experience and research. If you buy those units good luck finding someone who will calibrate.

It's a tough market out there with many of the older units still working fine. These bench meters get more stable as they age. You gotta make a living too you know.
 

alm

  • Guest
Re: Fluke 8846a teardown
« Reply #18 on: November 15, 2011, 02:46:20 pm »
Keithley makes some decent stuff as long as you avoid the 2100. The other (older) ones like the model 2000 are all their own design as far as I know.
 

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 802
  • Country: au
Re: Fluke 8846a teardown
« Reply #19 on: November 16, 2011, 10:50:26 pm »
Yes Keithley, I almost forgot about them. Famous for their nanovoltage measurements. I have a Keithley 2000, the internals looks very similar to HP34401. Down to the low themal emf relays that were used. Makes me wonder if they used 34401A as a reference. But the 2000 were not as robust as HP offering, it is also a lot lighter, Keithley does not have as many custom IC. Mine has a broken ohms function which is quite common, I have yet to fix. It holds calibration down to the last digit when compared to a freshly cal HP. It was calibrated in 2004, very impressive I wonder if it's due to age or design. Another downside of 2000 is that I can't turn off the display through the menu, I think it can be done through GPIB.

Regarding the cost, have you compared how much markup oscilloscope on prices? I am not to dwell into that. Fluke has a lot of experienced engineers to feed. Thus they also lead to a lot of patent filing. They have cal lab which is powered by a Josephson voltage standard, necessarily for low drift low uncertainty reference less than 0.001ppm. All these facilities cost money to run, and where does the money come from? Not forgetting the board of directors who only cares about profits. I could go on and on but I hope you get what I mean. Now I am curious, why are you looking at the cost based on the hardware? Why don't you compare it to a piece of 1 carat diamond? Does it worth more now?

If you feel you are being ripped off, try looking into a Fluke 732B, it only has two DC functions a 10v and 1.018v reference. Accuracy cost money and when you compare 8846A to the fluke/datron 8Digit 8508A, the 6digit 8846A is actually a bargain! On top of that, lifetime warranty is really good deal.

"10ohm to 1 Gohm with up to 10uOhm resolution" - Chinese friends huh...
« Last Edit: November 16, 2011, 11:17:23 pm by nukie »
 

Offline WartexTopic starter

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ca
    • http://headsplosive.com
Re: Fluke 8846a teardown
« Reply #20 on: November 17, 2011, 12:13:26 am »
Yes Keithley, I almost forgot about them. Famous for their nanovoltage measurements. I have a Keithley 2000, the internals looks very similar to HP34401. Down to the low themal emf relays that were used. Makes me wonder if they used 34401A as a reference. But the 2000 were not as robust as HP offering, it is also a lot lighter, Keithley does not have as many custom IC. Mine has a broken ohms function which is quite common, I have yet to fix. It holds calibration down to the last digit when compared to a freshly cal HP. It was calibrated in 2004, very impressive I wonder if it's due to age or design. Another downside of 2000 is that I can't turn off the display through the menu, I think it can be done through GPIB.

Regarding the cost, have you compared how much markup oscilloscope on prices? I am not to dwell into that. Fluke has a lot of experienced engineers to feed. Thus they also lead to a lot of patent filing. They have cal lab which is powered by a Josephson voltage standard, necessarily for low drift low uncertainty reference less than 0.001ppm. All these facilities cost money to run, and where does the money come from? Not forgetting the board of directors who only cares about profits. I could go on and on but I hope you get what I mean. Now I am curious, why are you looking at the cost based on the hardware? Why don't you compare it to a piece of 1 carat diamond? Does it worth more now?

If you feel you are being ripped off, try looking into a Fluke 732B, it only has two DC functions a 10v and 1.018v reference. Accuracy cost money and when you compare 8846A to the fluke/datron 8Digit 8508A, the 6digit 8846A is actually a bargain! On top of that, lifetime warranty is really good deal.

"10ohm to 1 Gohm with up to 10uOhm resolution" - Chinese friends huh...

8846a has 3 year warranty, not lifetime. They pulled free firmware from US site (dug up on EU website). Repair is $750 FLAT, even if you have a dead VFD or a cracked jack, it will be $750. FUCK THAT with a telephone pole. If the meter was ~$750-$900 retail it would be totally worth every penny, but at $1400 with flat rate repair - no thanks.
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Fluke 8846a teardown
« Reply #21 on: November 17, 2011, 02:44:01 am »
"10ohm to 1 Gohm with up to 10uOhm resolution" - Chinese friends huh...

And It works internaly at 10 nanoOhm mathematical resolution.  You can see it in the trend plot readings and in the manual offset function.

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 802
  • Country: au
Re: Fluke 8846a teardown
« Reply #22 on: November 20, 2011, 12:51:48 am »

8846a has 3 year warranty, not lifetime. They pulled free firmware from US site (dug up on EU website). Repair is $750 FLAT, even if you have a dead VFD or a cracked jack, it will be $750. FUCK THAT with a telephone pole. If the meter was ~$750-$900 retail it would be totally worth every penny, but at $1400 with flat rate repair - no thanks.

Calm down man, nobody forced you to buy the 8846? Okay I was wrong about the warranty and yeah I agree anything cheaper than the selling price is a good thing but you have no choice when the market command for such price. There's the Maybach and there's a Chery both will get you from A to B but I believe you already know you are gonna have to fork out a lot of money before setting a foot in the showroom?

If there's a true competitor out there which is more economical and perform just as well, they will probably have to readjust the prices.
« Last Edit: November 20, 2011, 12:53:39 am by nukie »
 

Offline WartexTopic starter

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ca
    • http://headsplosive.com
Re: Fluke 8846a teardown
« Reply #23 on: November 20, 2011, 04:56:59 am »

8846a has 3 year warranty, not lifetime. They pulled free firmware from US site (dug up on EU website). Repair is $750 FLAT, even if you have a dead VFD or a cracked jack, it will be $750. FUCK THAT with a telephone pole. If the meter was ~$750-$900 retail it would be totally worth every penny, but at $1400 with flat rate repair - no thanks.

Calm down man, nobody forced you to buy the 8846? Okay I was wrong about the warranty and yeah I agree anything cheaper than the selling price is a good thing but you have no choice when the market command for such price. There's the Maybach and there's a Chery both will get you from A to B but I believe you already know you are gonna have to fork out a lot of money before setting a foot in the showroom?

If there's a true competitor out there which is more economical and perform just as well, they will probably have to readjust the prices.

Borked 8846a made me depressed so to entertain myself I got an Agilent 34405A and a Micronix 2658 8.5 GHz spectrum analyzer. Early fucking Christmas :P
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf