Products > Test Equipment
Agilent/Keysight Acqiris U1081A Acquisition Card Teardown
(1/2) > >>
HalFET:
I managed to get my hands on a really rare, outside of industry at least, piece of kit this week: an Agilent/Keysight Acqiris U1081A (AP200). Couldn't find any close up pictures online, so decided to make some. As a small side note: I haven't actually been able to test it, on account of the exuberant fan duct bouncing into my graphics card. I currently don't have any thermal grease, so I can't take it apart (yet). But rest assured that's coming! For now already a sneak peak though. These seem to be sufficiently rare that you can't even find a proper datasheet on the U1081A (AP200). Luckily the one of the U1082A is quite similar, and the user manual contains the full specifications.

It has two input channels and an external triger, and will do 2 GSa/s with 500 MHz of bandwidth on one channel with 500 mV full scale, a maximum 2V offset, and possible 25 MHz . While inherently 8 bit, they can use averaging (there's more behind it) to increase this to 12 bit. It's capable of holding 256k 8 bit samples, and it can do quite a few fancy averaging things - for which I still have to read the manual. Trigger response time is about 18.8 ns. So in short, it's pretty much an oscilloscope (minus the frontend) in PCI card format. It sucks a mere 56.6W out of your computer, and is fully capable of (ab)using DMA to ship the samples off to greener DRAM pastures.

PCB Backside:



Top side (heatsink):



Side view:



Back panel:



Macro images of the back side:








More coming later!
HalFET:
Some information on the components so far:
Golledge 560T - TXCO - http://www.golledge.com/docs/products/tcxos/gtxo560.htm
Micrel SY100EL29VZG - Dual Flip Flop (related to clocking) - http://www.mouser.com/ds/2/268/sy100el29v-778210.pdf
HC595 - 8 bit shift register with tri-state outputs - http://www.ti.com/lit/ds/symlink/sn74hc595.pdf
DAC0800 - Differential current mode DAC - http://www.ti.com/lit/ds/symlink/dac0802.pdf
ST 93S66WP - Protected EPROM (couldn't find datasheet)
ICS570 - zero delay buffer - https://www.idt.com/document/dst/570-datasheet
LVT2244 - Octal buffer - http://pdf.datasheetcatalog.com/datasheet/nationalsemiconductor/DS012170.PDF
GSI GS8160236BGT-150I - 16 Mb synchronous SRAM (probably sample memory) http://www.gsitechnology.com/sites/default/files/partnumberdecoder.pdf

Spotted a couple more of the GSI chips on the non-visible parts of the board, so that's probably the acquisition memory.
alm:
Thanks for the pictures! Interesting how (at least for the U1082A) they added the FPGA with various firmware options (and possibly even custom firmware) for data processing faster than the PCI bus allows, without having to design a completely custom data acquisition system. I could definitely see the value of this for some applications.
HalFET:
So got a computer that'll fit this thing, which is good. Now comes the software dependency nightmare  |O

Also these are the most awful fans in the history of mankind, they make this high-pitched whining noise. And even with these fans the card runs very hot apparently. Might have to rig up my own cooling for this one!
maxwell3e10:
I recently got the same card and ran it with Agilent Acqiris software version 4.2 that is available from Keysight website. It works fine on Windows 10 machine, except for one problem. In AcqirisLive software no waveform shows on display if using averaging mode, although it does seem to be working. I can save the data in Raw ADC counts and it makes sense, except the number of waveforms averaged is reported as zero. I believe this causes the averaged waveform on display to go to infinity.

I am curious if anyone encountered such a problem. It is rather annoying since the point of the card is to display averages in real time.
Navigation
Message Index
Next page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod