EEVblog
Вставка
- Опубліковано 3 чер 2013
- Teardown Tuesday
Inside the Keithley 2015 THD 6.5 digit bench multimeter.
Not just a multimeter, it does low noise signal generation and automated audio THD system measurement.
Forum Topic: www.eevblog.com/forum/blog/eev...
Service Manual: www.tequipment.net/ProductImag...
Photos: www.flickr.com/photos/eevblog/...
EEVblog Main Web Site:
www.eevblog.com
EEVblog Amazon Store:
astore.amazon.com/eevblogstore-20
Donations:
www.eevblog.com/donations/
Projects:
www.eevblog.com/projects/
Electronics Info Wiki:
www.eevblog.com/wiki/ - Наука та технологія
Former Keithley engineer here, I was the lead application engineer for the 2015 for a while and later worked on firmware for this model and many others. Yes, the 2000 used the same main board. The same interface that communicates with the DSP board actually goes to the internal scanner card slot on the 2000, hence the reason the block diagram still shows the scanner card. The 2015 was actually developed for testing audio quality on cell phones. At the time, folks like Motorola and Nokia were using expensive, high end distortion meters that were overkill for testing cell phones, so we came out with the 2015 since they still needed a DMM for doing voltage and current measurements. It's not ultra precision for professional audio equipment, but enough to determine if a cell phone speaker is reproducing well enough or if it has a defect that's causing significant distortion.
Thanks a lot for your input!
I'm the really happy owner of one of these awesome instruments. I had to save up for a while and wait for a decent deal...
I'm a huge fan of Keithley instruments.
I worked for a major manufacturer back in mid 2000's and wrote a program to automate the calibration of the 2015 Keithley meter. Turns out the factory firmware had a bug which I found (pat on back). The new lower resistance range calibration constants were lost after powering down the 2015. It was not apparent unless it was powered off and retested. It took me a long time to notice it was happening because powering the unit down after testing the accuracy was never a part of our manual calibration procedure and they rarely failed. When I decided to automate the calibration on our Datron 4808 system I used a single Keithley 2015 meter as my guinea pig and noticed a problem after my first "successful" write of the calibration constants with my new automated program. On the second try the meter came back failing on the lowest resistance range which I knew I had written and passed the day prior. What was occurring was after power down the newly written calibration constants were lost and the old ones still remained. We must have calibrated dozens of meters manually and never knew what was occurring until it was automated. Luckily our factory lines had documented the ranges which were needed to test product and that range was not used. In other words product was unaffected which could have effected millions of devices. Keithley technicians provided us with about a hundred free eeprom chips to correct the issue.
I'm saving this video for "work" tomorrow. I love teardowns!
Oh man, pre-ROHS reflow soldering in high quality gear just looks so spectacular when it's done properly! The best lead free reflow in the world doesn't look that nice and clean and shiny. This is a beautiful piece of instrument Dave, and this is the best video you've done in months in my opinion at least.
Peace from NZ.
Chris
I love those old metering devices! Thanks Dave!
Another 36 minutes entertainment, thanks Dave.
Dave I’ve got 2 of these excellent DMMs one is for backup the other one is on the test bench and its also very handy for accurate Sinad measurements on the radios I work on. These still bring very good money in the USA.
One of the great things about the 2015 is the fact that it can make 4wire resistance measurements up to 100 MOhms. Most other Keithleys stop at the 2 Mohm or 10 MOhm range.
Also, that retention clip is much more than a retention clip. There is a ferrite clamp-on core in there, to help shield the wires. lol guess I spoke too soon :) not 5 minutes later you found that out!
Surprised not to see any fancy caps near that ADC for the dual slope integrators. Nice teardown, great meter!
You are the best Dave!
Thanks for another awesome teardown. Thumbs up !
Looks like it's got a gas discharge tube in the input protection as well. Pretty sweet piece of equipment.
Ok, know this was done some years ago, but, note: common failure on these is the front/rear input switch :)
Hey Dave, the LM399 Voltage reference always has a wobbly case. I have a few of them and they are all wobbly... and I am building a reference with it. Its the heat insulating case.
Dave Love your videos to bits , I have a 2410 here that starts but has some error affecting all tests , shall soon be going up on ebay to who ever wants to have a crack at it , as I will and can not bring myself to break the cal stickers
Hello! I have a 2015 for years and the problem has always annoyed me.sometimes when I power it up, it beeps with no display, paritial display..all smudged display or getting lucky, after the 2nd or 3rd try..or more it works and stays on without any additional problems...until I turn it off again.
This one is on my Bucket list. I want it for the THD function because I construct stereo amplifiers as a hobby.
Well theres much more parameters than just THD in amplifiers (I'm sure you know). Why dont you save up for a used audio precision or a dScope?
I used to work for Keithley, and I was the lead applications engineer for the 2015 for a while (around 1998-1999) before I went into design engineering and worked on the firmware. While it is a THD meter, it wasn't really intended for high end audio, it was targeted for doing audio quality measurements for cell phones. I was curious once and borrowed one to see how my 1960s vintage console stereo (all tube) fared, and to be honest, it was at the limit of what it could measure. Don't quote me, but I thought the 2015 could only accurately measure down to about 0.3% THD. Adequate for a cell phone, not so great if you're trying to test high end audio.
Thanks for the video. I see it has a Math function. Does that mean it can be set to calculate audio wattage?
Moving average algorithm:
lastVoltage += (currentVoltage - lastVoltage) * updateRate;
The above can also function as a low pass filter if applied to an audio signal :)
@Richard Smith Great hint with the low pass filter. I use the moving average function a lot, but it never crossed my mind that I could use it that way for audio signals. I will experiment with this LPF in MaxMSP later tonight.
I'm not sure I get this guard trace thing. Is it anything to do with leakage? Preventing leakage into certain pins? I might need it explaining a few more times ;)
Your tilting bale is on upside down...but yeah, they stink.
Source: I used to work there. And regularly ripped them off any gear I was working on.
Hi Dave.
I am looking for a 2015THD. Do you still have it. Would you sale it? Please let me know. Greets Form Austria to Australia
Not yet. The odd leftover screw of course...
I just thought of a great fundamental friday topic: Guard VS Shield :)
No doubt it's high stability, probably a few ppm at most. Just that I have never seen a 1% version of those high stability metal film ones before in that package, and marked with 70K000 to boot.
Hi! Is there a way to download all the original pictures per project from flickr at once goal? I have watched some of your video I like the energy, sometime I wish you could go in more details specially in the teardown, I was going to suggest that make some high resolutions pictures until I saw this video, I am interested on learning layout so I suggest make pictures with lights underneath to see the tracks would be just perfect. As you I am interested on learning bit to take into my designs!
What voltage you got in austalia
The reason for those resistors to be glued together so their drift will be similar as they get heated together to the same temperature.
I think.
Usually industrial-type devices like this aren't sensitive enough to be damaged. (From a build perspective, not precision). It would be a good precaution though. Where I live it's too humid to give anything enough of a shock to cause damage, I'm not sure how it is in Australia. I work on high-end computer systems all the time for work and I don't even own a strap.
Aaand, the price of Keithly stuff on eBay spikes for the next month.
Where do we apply to be your camera operator?
If you look realy realy close at C225 you cold see Dave's reflection in it's solder joint :)
PS for once I'd actually be interested to see how the THD measurement features work / are operated. It is quite unique.
Awesome!
you were sayin about camera operatorship, thay wouls be a DOP or director of photoreceptor. also in a way i freal a bit sory that its not a robertsons screw to un do. i know there knormally for wood and decking and frames but being a canadian my self i have to say there darn nice to work with
Was walking home in caringbah the other day and saw a bright orange van with "david jones electrical" or similar. That wouldnt have been you would it?
For $4300 US these had better be well made & well designed. These are dynamite units. I always have my eyes open for these at a good price. These may have "obsolete" parts and be getting long in the tooth, but Newark is still selling them. The THD feature is nice. I used 4 of these with a Sony built Tektronix 371B curve tracer which was used by day to test transistors for 3 mfgrs. In the evenings I used the 371B to test vacuum tubes. I am using Fluke DMMs now & one 2015 with an ancient Tek 576.
It's a system multimeter, they need to be that fast for automated system measurements.
70K 1% may be high stability - no need for high absolute accuracy if you are software cal'ing - just good stability
I snagged one of these as non-working for a few hundred on ebay It just needed a new fuse
yes youre right but its diferent from an moltimiter
try 240 should be fine UK im gessing?
Yeah, but a 100uA range would have covered down to nanoamps! Remember this is Keithley we are talking about here.
Did I see a chip in there marked "National Instruments"?
All those optocouplers are Agilent/HP devices which have been around for ages...if it ain't broke don't fix it!
On the flip side they will argue that high voltage supplies with low current will not kill, so that "proves" that high current kills, even at very low voltage. What they do no understand in this case is that the high voltage power supply has high internal resistance that gets the majority of the voltage drop when a human becomes the load. Therefore, the output voltage greatly drops which in turn is not high enough to move enough electrons (current) in the human body to stop the heart.
For fundamental Friday you should do an explanation on the most misunderstood concept in electricity that people go around talking about how current kills, current kills, blah, blah..... even if they think you touch a 12 volt battery. What they ignore is that voltage potential needs to be high enough to get electrons in the human body to move because the bonds are much stronger, being that we are more insulating than conductive like that of copper. Continued....
how is it doing 6 digits with only 16 bit adc?
+Mioara Voicu 16bit ADC for the THD measurement digitizer. The 6.5 digits come from a delta-sigma converter in the top board
I saw a department store called David Jones - that wouldn't have been you would it?
An Android tablet doing the datasheet? Awesome!
There's precision multimeters and then there's precision multimeters.. wow.
One has THD measurement, the other doesn't. Not trying to be smart, that's the difference. This is basically the only meter on the market that measures THD.
You might as well have a crack at it, as the cal stickers are worthless if it has a fault.
it probably wasn't the best explanation though!
plese tho a computer terdown
Should you set 240V or 220V if your mains voltage is 230V? :>
240
Does Anyone sale a 2015 THD
Pleaase! Read out those ROMS!
whats the device with the FOX logo i also saw one in the apple macontosh dave took apart
A tri state oscillator.
thank you!
Dave, why are you complaining about no microamp range? It has milliamp range, which in turn has 5 or six decimal digits - this covers even tenths of microamps!
"bit of a hodgepodge" :)
Finally, one way to show them is to take two, 100 mega ohm resistors and connect them to the 240 volts mains. Then measure it with a meter that is high impedance input. It will read 240 volts. If you touch the opposite leads on the two resistors then the 240 volt mains will not kill you (only professionals like you and me should do this!!!). However you then show them that by taking away the resistors you again get 240 volts showing up. But this time you sure as hell will not touch it!!
My supply is 240V
-65dB residual THD is a bit disappointing. Even sub-$100 computer sound cards can do -95dB these days.
My oscilloscope have a tilting bail like that, and if you want to take it apart (which you do), you need to pull those adjustment things on the sides, while pulling the insides forward... so you need four hands do take it apart decently... dumb shit
I can only find it from 4,0000$-5,000$ used on ebay!
I found one for $339 on eBay...company sell-off. No issues with it at all thus far...had it two years now.
Total Harmonic Distortion
Disappointed, you turned it on first.
Getting moist. lmao!!
Electrical - what's that? I do electronics!
You didn't care to watch till 1:25 ?
you should teardown an xbox 360
djelectricians[dot]com[dot]au