Oh very interesting, thank you
As expected, the ADI implementation of the LTZ1000 circuits are flawed, they do not conform to the original design and does not produce the original specifications. I would not recommend buying this board if you really want a good LTZ1000 reference. They don't know what they're doing yet, they have no experience with this design and should have kept to the original design of LT. They used the wrong resistors too....I know you were expecting me to say it but it is a fact none the less.
Yeah had a quick look at the design. I must say that i am too lazy to build one myself though, its time consuming..
I contacted them already, found it an hour or so ago too.
I will let you all know the price when they respond.
edit: IslandLabs guys say 245 eur for LTZ1000ACH board
edit2: the friedrich-messtechnik guy said they only sell to companies
Having bought two from island labs I will say they seem to be good quality. When purchased you get data on aging, noise, current calibration, bom, and schematics. They're also fairly helpful with questions and offer 1 free calibration(I forget the time span).
Attached a couple of pictures I took last year of the LTZ1000A board. If I have time I'll take more detailed ones later of both but since they're in two boxes and one is soldered I don't have time just now.
It just wouldn't be fun not build some of my own... I am interested in converting this into a non-LTZ reference though. That's why I asked about best way to safely cut PCB's... IDK about tin snips but I know it's been done. It's in a working 8842A now. Somehow the reference might be more fun than the old MM to me I think.
Bill
It just wouldn't be fun not build some of my own... I am interested in converting this into a non-LTZ reference though. That's why I asked about best way to safely cut PCB's... IDK about tin snips but I know it's been done. It's in a working 8842A now. Somehow the reference might be more fun than the old MM to me I think.
Bill
Didn't you build an enclosed 10V reference with this method @Micke?
Yepp, I had a 8842A DMM beyond rescue...
I just used a hacksaw to cut out RefAmp, thought of using Dremel as well, but thought that hacksaw would generate less vibrations, and easier make a straight cut.
Could have made RefAmp board even smaller, but wanted to be on the safe side, and to have somewhere to put mounting screws...
I soldered power +/- 15V directly on Op-amp pins U702 on the secondary side. For Ground, I used the via holes just under the text "C103" (think shield was soldered in these holes originally)
I am not very impressed by the stability though, typical run with FLUKE 8846A below (PLC100), SZA263 in metal box with lid on, powered 24/7 for weeks:
* Min 10.00005V
* Max 10.00010V
* Avg 10.00007V
* SD 7,6µV
* 40000 # samples
So a span of 5ppm between Max and Min, did expect better actually!
So a span of 5ppm between Max and Min, did expect better actually!
SZA263 is a zener with a transistor on the chip. You cannot expect "LTZ1000 stability" unless the module is ovenised, imho.
[...]
* Min 10.00005V
* Max 10.00010V
* Avg 10.00007V
* SD 7,6µV
* 40000 # samples
So a span of 5ppm between Max and Min, did expect better actually!
5ppm over several weeks is not too shabby? Is the area temperature controlled?
[...]
* Min 10.00005V
* Max 10.00010V
* Avg 10.00007V
* SD 7,6µV
* 40000 # samples
So a span of 5ppm between Max and Min, did expect better actually!
5ppm over several weeks is not too shabby? Is the area temperature controlled?
Did not record start/stop time, but 40000 samples at PLC100 is unfortunately only like 22h
Yeah, did not expect LTZ1000 performance, but I am pretty sure the SZA263 drift more than my AD581LH (need to find the notes from the measurenents...
)
I've finally started logging voltage and temp together. I'm using 1 array of 10 DS18B20's to measure ambient, 1 array of 5 for each of 2 meters. The placement for the arrays for the meters is essentially placed between the ADC and the reference. Is that the best place to measure the internal temp, or should it be more central to the analog board?
Hi,
I tested Analog Discovery 2 spectrum analyzer function together with pipelies LNA (0.1 ... 10Hz) and found that this is a decent setup, without the need of a boat anchor on the desk. Thought that could be of interest for you guys. As an example I measured LNA with shorted input, a 9V battery and one of my LTZ references.
-branadic-
[...] together with pipelies LNA (0.1 ... 10Hz) [...]
What is the "pipelies LNA" ?
I always wondered how the noise spectrum of my LTZ references would look like. Sure we can use DMM readings to get a raw idea, but even the 1/f region is always hard to get. Now that I have one of those nice Digital Spectrum Analyzers (Advantest R9211) or as they are also called Dynamic Signal Analyzers or FFT Analyzers, I was acutally able to create such a nice noise plot today.
-branadic-
I always wondered how the noise spectrum of my LTZ references would look like. Sure we can use DMM readings to get a raw idea, but even the 1/f region is always hard to get. Now that I have one of those nice Digital Spectrum Analyzers (Advantest R9211) or as they are also called Dynamic Signal Analyzers or FFT Analyzers, I was acutally able to create such a nice noise plot today.
-branadic-
That is very cool! - what is the noise floor of an instrument like that?
Hi,
I tested Analog Discovery 2 spectrum analyzer function together with pipelies LNA (0.1 ... 10Hz) and found that this is a decent setup, without the need of a boat anchor on the desk. Thought that could be of interest for you guys. As an example I measured LNA with shorted input, a 9V battery and one of my LTZ references.
-branadic-
An LTZ1000 with no flicker noise?
Mhm,
I can explain the 50 Hz peak in the spectrum: your cookies box isnt tight enough.
But where do the 60 Hz come from? (Screen frequency of the instrument?)
with best regards
Andreas