I bought a Samsung SSD a few days ago. I checked tracking yesterday morning and saw that DHL had handed it off to USPS with an estimated delivery of today (Friday). I checked tracking again today and saw this:
Your item was returned to the sender on
August 12, 2016 at 10:46 am in City 2 (30 miles away)
because the addressee moved and left no forwarding address.
Date & Time August 12, 2016 , 10:46 am
Status of Item Moved, Left no Address
Location City 2 (30 miles away)
Date & Time August 12, 2016 , 4:29 am
Status of Item Arrived at Post Office
Location City 2 (30 miles away)
Date & Time August 12, 2016 , 3:14 am
Status of Item Accepted at USPS Destination Facility
Location City 1 (where I live)
Not sure what to make of that. Apparently it arrived at the local post office, but then was returned to sender from another post office 30 miles away a short while later. And I have not moved in the last 10 years.
edit:
It's now Saturday morning and magically, my package is "Out for Delivery".
I received a BK 1856D that I ordered from eBay ($175) a couple of days ago (seller was nice to ship quickly and priority instead of economy).
So far I am really pleased with it. I calibrated its internal TCXO to my rubidium reference (RFTG) without any trouble.
It has the option of using the internal clock or an external one, and if you enable the external one, it will display 1 more digit. I wish they had just displayed all digits even when the internal one is selected even if it bounces around though. I may try to find out where it "tells" the microcontroller (Atmel AT89C52) which one is enabled and find a way to override it. It does take a full 30 minutes for the internal TCXO to warm up just like the manual says.
The display shown in the picture is it using the Rb clock as an external clock and me feeding what my SDG2082X "calls" a 10 MHz signal into it. It is rock solid and does not bounce around, the SDG2082X starts up cold around 0.1 ppm off and then warms up to 0.16-0.17 ppm off. I am pretty impressed with the Siglent generator, it is one of my favorite toys.
I also tested a 26 MHz Pletronics OCXO I got off eBay and it bounces around much more than the SDG2082X in the final couple of digits. Maybe it would do better if it was actually soldered into a pcb instead of just the old breadboard it is in.
Alan, glad to hear you're liking your 1856D. They don't show up often on eBay. I have one in my repair queue that, at first glance, appears to have an issue with its power supply. Good to know the TCXO is friendly to adjust.
Alan, glad to hear you're liking your 1856D. They don't show up often on eBay. I have one in my repair queue that, at first glance, appears to have an issue with its power supply. Good to know the TCXO is friendly to adjust.
The power supply doesn't have a whole lot to it, hopefully you will be able to fix it easily.
I thought I found the way to enable the extra digit - seems to be controlled by the CN6 connector on the main board. It is shorted for "external" and open for "internal". I tested the input/output clock jack to make sure it didn't act differently to see if the microcontroller would switch it somehow and the input/output looks the same on it whether CN6 is open or shorted. I thought shorting it would enable the extra digit, but I'm not so sure it does. The last digit seems to come up "0" a bunch, more than you would think it should. Not always though, when I feed a signal from my AWG, I can alter it and have the last digit be non zero, but using its internal clock and monitoring my 10 MHz reference signal, it went from 60, to 70, to 80, for those last two digits and I wouldn't have expected that. You would think if both the internal and external clocks are 10 MHz that it wouldn't matter but perhaps it does more than just turn on / turn off an extra digit...
Well, maybe it works fine, the difference with the 60-70-80 seems to be when it is below 10 MHz, even with the external clock. Perhaps the "CN6 short mod" is the way to enable an extra digit. It doesn't bounce around even on the internal clock....
edit : it seems that whatever algorithm they use loses a digit when the first digits are 99999, other than that you do get an extra digit by shorting the 2 pins in the CN6.
edit2: I guess it really isn't losing anything. In 10s mode, log10(10000000*10) is 8 digits and that is the best it can do. log10(9999999*10) is 7.99999 digits.
Interesting. Maybe start a thread about this counter. Last I looked, there didn't seem to be much about it on the interwebs.
A bit of an unknown but for the price AUS$70 who could resist! Its local so no post.
A bit of an unknown but for the price AUS$70 who could resist! Its local so no post.
Great condition! I certainly wouldn't want to ship it.
MS Office Home & Business 2016. After years of believing in and advocating open source software, I finally decided I can not live with LibreOffice's bugs and sluggishness, and the saved $250 on software simply can not make up my swearing, misery and time loss.
Same. I cannot deal with Open/Libre. So little time in the day to fiddle with software.
Today I got a Rohde CMD55 GSM800 and 1900 tester loaded with options, even OCXO, audio analyzer...
Curious what that instrument is used for. Do you have a plan for it?
Well, it has tons of functionality to test gsm mobiles, and besides that 2 rf generators, audio codecs and interestingly a current/voltage/power digital meter. Actually I bought it cause Im interested in seeing the internals and getting to know the audio part but eventually I'll probably sell it in this side of the pond
Same. I cannot deal with Open/Libre. So little time in the day to fiddle with software.
The thing that triggered me is last Friday when I was preparing an important report PPT for TI, LO crashed several time, and that makes me to save multiple times per minute, fearing I will lose more data.
What's worse is, all LO instances share the same process, so if one crashes, all other instances crash at the same time. I believe implementing SLSJ or SEH won't be hard, but they simply does not want to do so.
Some commercial tools, on the other hand, have moderate data loss protection. For instance, Altium Designer will catch SEH violations (aka. SIGSEGV in POSIX), then pop up a window, and allow you to continue working.
I feel very happy and privileged that all my employers didn't mind me being a bit of computing greybeard and using LaTeX beamer package to make presentations.
Altium Designer is a Ferrari of design suites when it comes to data loss protection. AutoCAD will gladly leak all memory then just crash, or sometimes will generate a design file that results with crashing the program when trying to open. Solutions on Autodesk knowledge base are just ridiculous - "reinstall video driver", "reinstall .NET framework", "create new Windows user account", "use Autodesk Crash Dialog on Close Hotfix" and so on.
I got this gem of a book. Power Integrity by Steven Sandler. I had met him at a trade show and decided to buy his book. It covers the tools and techniques to measure and analyze modern power systems. It is a tough challenge and this book clears up a lot of mystery for me. It also has me scouring eBay for more test equipment. The book was cheap. The resulting equipment purchases are not so cheap.
Picked up a 10G ohm 10kV voltage divider. It was bit of a gamble as Unilab is an educational equipment brand but opening it up revealed some decent quality Welwyn 2% resistors. I bought it because of the difficulty of finding 9G, 900M etc resistors.
Initial testing indicates that the ratios are actually pretty good and I will only be using it where I want to measure higher voltages with higher than 10M loading. The 1kV and 10kV terminals have additional insulation. The cable was connected across the 1M resistor but I've removed it and will either fit a more convenient one or just use the terminals.
Quiz: How many errors can you find in the front panel text?
Picked up a 10G ohm 10kV voltage divider. It was bit of a gamble as Unilab is an educational equipment brand but opening it up revealed some decent quality Welwyn 2% resistors. I bought it because of the difficulty of finding 9G, 900M etc resistors.
Initial testing indicates that the ratios are actually pretty good and I will only be using it where I want to measure higher voltages with higher than 10M loading. The 1kV and 10kV terminals have additional insulation. The cable was connected across the 1M resistor but I've removed it and will either fit a more convenient one or just use the terminals.
Quiz: How many errors can you find in the front panel text?
10gohm is too much for "only" 10kv, I have bleeders used in the >100kv range at around 500Mohm... a problem with that impedance is that bandwidth gets reduced to a few Hzs and bandwidth here means that at higher freqencies the output gets higher not lower...
Anyway good buy for the strange resistors
The book was cheap. The resulting equipment purchases are not so cheap.
Similarly, visiting this forum is free...the shopping that results, not so much.
Similarly, visiting this forum is free...the shopping that results, not so much.
Ain't that the truth!!!
<edit to add - I ordered a used copy of the blasted book from Amazon, too, because it looked interesting.
>
-Pat