Products > Test Equipment

Agilent E7495 linux root account

<< < (24/91) > >>

DogP:
Here are a bunch of pics from my battery pack rebuild.  It's a 9 cell 18650 pack, with SMBus, fuel gauge, protection, etc.  I'll try to go through the process, and mention any gotchas as well.  But first of all... let me start by saying I've never rebuilt an 18650 battery pack, and I don't own a battery tab welder.  So, I bought batteries with tabs already attached, so I could solder to them (soldering directly to 18650s is hard, and the heat isn't good for the battery).  The battery tab welder would have made things quicker and easier though.

The obvious things: safety first... lithium batteries hold a lot of charge and can be dangerous, catch fire, etc.  Pay attention to battery polarity, and don't attempt this if you don't know what you're doing.

I chose the HyperPS brand batteries, with tabs attached, which were <$3 each shipped on ebay (these: www.ebay.com/itm/272359448176 ).  They claim guaranteed >2600mAh, but suspicious of unknown brands, I made sure to test them in my BT-C3100 charger/tester before building the battery pack.  These actually tested quite well, >2800mAh each.  In case you don't know, fake 18650 cells claiming ridiculously large capacity (6000mAh, 9800mAh, etc) are all over the place, and generally they are actually junk (<1000mAh).  A real 18650 will generally be around 2000-3000 mAh.  The original batteries were MOLI FSPE.70037 (ICR-18650G), which were 2000mAh cells.

GOTCHA: make sure the length of the new cells are close to that of the original cells (~65mm).  The original cells are unprotected, so you'll probably want to stick with unprotected cells, as the protection circuit generally adds a little bit of length, and there isn't a lot of extra room.

To open the pack, peel back the short side of the label (or cut it, though if you leave it in one piece, it works well to hold the pack together).  Go around the edge of the pack with a knife to crack the small dots of superglue holding the plastic pieces together at the edges.  You can also crack the glue by twisting it like you're popping ice cubes out of a tray.

GOTCHA: the pack was assembled in the bottom half (side with the battery contacts), and then some RTV was put on top and closed up.  You want to disassemble it in the reverse order.  If you pull the pack out of the bottom (like I did), it'll bend the battery contacts and rip the LCD cable (I ended up replacing it with standard 0.05" ribbon).  Instead, you should open the pack by prying the top from the batteries.  Then, once the top is open, lift the pack straight out of the bottom.  Clean out any large globs of RTV from the case so they don't interfere with the fit of the new cells.

Once it's out, desolder the two tabs from the PCB, undo the Kapton tape holding the fuses, and cut the metal strips connecting the fuses to the batteries (leaving as much of the strip as possible, as you'll probably want to reuse it), and remove the PCB and fuses.  Once that's removed, you can pop apart the original pack and see how it's connected... it's 3 series packs of 3 batteries in parallel.  You'll want to start by assembling your 3 parallel battery packs.  If using the solder kind, rotate the tabs so you're soldering away from the center of the battery, so it'll solder easier, and apply less heat to the battery.  Make sure the tabs also line up on all packs, so they're easy to connect in series.  The battery pack is tight, so try to keep the solder fairly low profile.  Once they're all assembled, drop them in for a quick test fit.

Next, you need to add the tabs that go to the PCB.  Look at their placement on the original pack, and try to get it as close as possible on the new pack, so the PCB will fit properly.  I added Kapton tape between the packs, so all current will flow through the tabs... though that's not very important, as they're connected in series anyway.  I then soldered the tabs between packs together and slipped a piece of heatshrink over them and taped them down with Kapton.  I then transferred over the tape from the old batteries to the new ones for the boards and fuses, and put a few pieces of Kapton over the pack to keep it held together.  I desoldered the red and black wires from the PCB (so the PCB was completely out of the way), and connected the fuse tabs to the tabs on the ends of the pack, and cut off any extra tails.  I then put Kapton over the ends, since there wasn't enough room in the pack for the cardboard endcaps that were on the originals.

This gives you batteries fully assembled, with a red and black wire and two tabs sticking up.  Then solder the red and black wires to the PCB and press it down over the tabs and solder the tabs.  Then carefully drop the entire pack into the bottom of the shell, making sure the battery contacts go straight into the slots.  You may need to pry the plastic back a bit on the LCD end to pop the LCD in place.  Then with a multimeter, make sure you have around 11V across the points where you soldered the red and black wires.

Finally, temporarily close the lid, slide it into the E7495, and make sure it's functional before permanently closing it up. Once you know it works, I'd put a few dabs of superglue around the plastic shell, pull the label tightly over the shell and seal it up (the label should still be tacky, but will probably need a bit of glue/adhesive for a permanent seal).

The battery fuel gauge will be wrong (mine said 0% when it started, even though it already had a half charge), but after a full charge and discharge, it figured itself out.  There's an option to refresh the battery on the E7495 (automatically charges, discharges, and recharges), though it didn't work for me (it never completed because the gauge never showed that it was fully charged).

Anyway, I hope this helps... or convinces you to just spend the $200 on a new one. :-P

Pat

TheSteve:
Nice rebuild of the battery. I need to find some cheap/free packs from the 7495's big brother the N1996A to rebuild. I rarely want to run it on batteries but once in a while it could be handy.

btw for anyone doing a battery rebuild like this be sure to start with all cells having an equal charge status.

smgvbest:
Thank you for the battery rebuild,   that will come in very handy.  I've got one battery going thru the recondition now and the other no so healty so this is great timing.

Also thank you for the hack to enable options.   I don't have the HW for Opt 500 or the E1/T1 options but every thing else enabled just fine.

Does anyone know how to get to a self test screen or if it even has one?

DogP:

--- Quote from: TheSteve on October 23, 2016, 12:30:47 am ---btw for anyone doing a battery rebuild like this be sure to start with all cells having an equal charge status.

--- End quote ---

Ah, yes... good call on the charge status.  Another note... according to the manual, the E7495 heavily drains batteries even when turned off.  It says 18% per day, just to monitor for the power button :palm: .

I know what you mean about batteries though... plugged in is nice on the bench, but laziness usually gets the best of me, and not having to find an outlet somewhere else makes me more likely to use it.  That's why I've got a battery powered oscilloscope as well.

Oh, somewhat related... I tested the batteries I pulled from the original MOLI battery pack.  One battery was shorted, which drained all 3 of those connected cells to 0V and ruined them.  The other 6 actually tested good, between 1700 and 1900 mAh.  So, I'll keep them around for future projects where a mediocre used cell will work fine.


--- Quote from: smgvbest on October 23, 2016, 01:20:31 am ---Does anyone know how to get to a self test screen or if it even has one?
--- End quote ---
Under system->pg. 2->service->verification there's a display and button test... but it's not much of a self test.  I'm not aware of anything like a memory, etc. test.

Pat

DogP:

--- Quote from: sportq on October 22, 2016, 12:16:55 pm ---Pat, really interested in the waveforms you've created. What are they? FM and AM would be useful but I'm not sure there would be any control over the modulation parameters.
--- End quote ---

For anyone interested in making your own waveforms, but not really sure where to start... here's a quick tutorial for making a compatible FM file in Octave (should work in MATLAB too).  You'll need the communications package installed (I believe you'll need the communications toolbox for MATLAB).

I'm using the piano2.wav file from this page: http://www.kozco.com/tech/soundtests.html as an example.  Below is the code... first you read the wav file, resample it from 48Ksps (audio file sample rate) to 250Ksps (iDEN file sample rate), and take just one channel of the stereo file (and rotate the matrix).  Then you FM modulate it to a complex baseband signal.  Then to get the signal ready to output, you need to scale it to full 16-bit integer scale, shorten it to 360000 samples (length of original iDEN file), and break the real/imag part of the complex signal into seperate columns of the matrix (for fwrite to output properly).  Finally, simply open the file for writing, write the file, and close it.  Then copy it to your E7495, put it in place of the iDEN file, and test it out!  The process is basically identical for the EVDO signals, except they're 2.5Msps, and only 200000 samples long.


--- Code: ---pkg load communications
piano=wavread("piano2.wav");
music=resample(piano(:,1).', 250, 48);
fm=amodce(music, 250000, "fm", 7500);
cp_fm(1,:)=real(fm(1:360000))*32767;
cp_fm(2,:)=imag(fm(1:360000))*32767;
myfile=fopen("fm_music.bin", 'w');
fwrite(myfile, cp_fm, "int16");
fclose(myfile);

--- End code ---

Of course the E7495 doesn't go down to the normal FM broadcast band (only goes down to 375 MHz), so I used an RTLSDR dongle with SDR# to demodulate the FM at a frequency it could work at, to verify it worked.

Pat

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod