« first day (1616 days earlier)      last day (3327 days later) » 

4:36 AM
@PlasmaHH For my wedding, one of my groomsmen got me a tube of 555's as a joke
@Mike looks pretty good - a bit of detection code and you should be in business
A thought: look at how long each rising and falling edge is (probably 5-20 samples?).
Once you know that, you can do a really simple level detection algorithm. For example:
If you have an array of 10000 samples, and you know that it takes less than 10 samples for one of the fast transitions, then I would pseudo-code it as:
x = data[0 to 9,999]
for t in range(9,989):
    delta = x[t] - x[t+10];
    if delta > threshold_1:
        pump_1_count++;
    if threshold_1 > delta > threshold_2:
        pump_2_count++;
Of course that will probably need some tuning to prevent multiple increments per edge, etc... etc...
 
5:36 AM
@Mike If you are using USB to power the thing, why not use a USB to UART cable and a little microcontroller that feeds some simple data straight into the tablet?
microcontrollers are largely software, that's your field of expertise ;o)
Ah I see you found Arduino and the likes in the mean while yourself. You don't need a full Arduino, you only need a small 8-pin controller for a job like this. Although Arduino (or a similar board like the other one you linked) is much easier to use.
 
 
2 hours later…
7:36 AM
@W5VO nice, I didnt get anything geeky for my wedding :/ the nearest thing I ever got was an atx power supply case as gift wrapping... it was one that almost went up in flames... the gift was nice too... a tshirt with "No I won't fix your computer" on it... made my time at university definetly more enjoyable...
hm, I wonder if user 70892 is a troll ...
 
 
7 hours later…
2:53 PM
Did anyone discover this year's April fool's on EE.SE yet?
 
3:27 PM
you mean it isnt the whole blog? ^^
 
 
5 hours later…
8:23 PM
argh, noise everywhere ^^
 
8:39 PM
@PlasmaHH If you want to see a troll-esque creature, look at this one. Look at its other SE accounts too.
 
8:52 PM
yay
@NickAlexeev Id rather dig through the manual of this scope ^^
 
@NickAlexeev But it's tagged , so it must be legit
 
ledit you mean
btw. is legit a legit word?
 
@PlasmaHH "legit" is just a slang shortening of "legitimate"
 
@W5VO I know, but there was a question on english.se about it that was first worded almost like that...
 
it's in the dictionary - must be a word ;)
 
8:58 PM
hehe
hm, I wish manuals came in a siri like format... "hey manual, whats this damn button for?"
 
@PlasmaHH just push it, see what happens
 
9:15 PM
bang
I need to get used to this ultravision stuff but it seems nice
 
9:27 PM
@PlasmaHH If it is the first DSO, the trial time of the options is too short in my opinion.
I like mine.
 
@jippie depends on your ethics for "extending" the trial time...
 
@jippie As Ringo said in Yellow Submarine: "I can't help it -- I'm a born Lever-Puller"
 
@ScottSeidman so ... what happened?
I was like 4 years old or so, last time I say Yellow Submarine
maybe 7
don't knwo
 
And I hope you didn't miss the pun of Liverpudlian!
 
I did miss that ...
@ScottSeidman But I get it now
Not native English ..
 
9:33 PM
@jippie I hear they don't love Ringo in Liverpool
 
@ScottSeidman don't know
 
Just got my Rigol DS1054Z Monday. I have a Tek DPO scope, that cost about 10X more when I bought it, but this thing is pretty good.
 
yeah, just playing with the 54Z here too
will probably get some other probes too since at least one isn't really usable
 
@ScottSeidman I had a chance to play with a demo unit and it was very reasonable.
 
@PlasmaHH Pretty powerful for the price point. I don't care for the measurement display that never really entirely clears until you reboot. Haven't checked out all the probes, but the two I've used are certainly functionable.
@W5VO It's also hackable to do some nice things, but I'm not going to do it.
 
9:38 PM
@ScottSeidman one of them looses compensation when you shake it ... but otherwise they are ok
 
I'm trying to mush through a Keithley SMU manual
I swear... I think it's running Linux underneath the hood
(or some OS)
 
@ScottSeidman what do you mean by "entirely clears"? all the rise/Fall etc. times on the bottom? you can remove them in the measurement menu
@W5VO I would bet most do
 
@PlasmaHH I bought it from Saelig, which is local for me, so I picked it up at their site. Chatted up a sales rep, maybe about retooling our student lab. Looks like I'd have a hard time spending a grand US on a student scope these days. Our 4-channel tek student scopes were about $1200 when I bought them in quants of 12!!
@PlasmaHH You can remove them, but say you put up three and remove the first two. They'll appear as greyed out on the bottom line -- not just gone. A bit annoying. My understanding is that the greyed out ones scroll off as you add more.
 
@ScottSeidman the new HP scopes can have autoset disabled! killer feature!
 
@W5VO Pretty sure this Rigol series lets you disable Autoset, too.
 
9:44 PM
@ScottSeidman ah that you mean. yes the greyed out get removed if you add more. I think it is greyed out because you can chose to recover. chosing F6/delete all will remove all and then you can add new ones "fresh"
 
@PlasmaHH I don't think that starts them fresh. Add three or so, then delete them w/ F6, and add one more. I think you'll see greyed out boxes.
 
@ScottSeidman oh, hm, indeed... well ... guess that would be ok for me...
 
@PlasmaHH Yeah, nothing to do w/ functionality, just a hair strange. I suppose I'll get used to it, then I'll think all scopes should work like that.
 
hehe... I am currently trying to get an idea under what circumstances it can measure rise times...
 
@PlasmaHH I gave up designing circuits with rise times. I just use fall times. I start at 10 volts, and only move down. I find that if I can't get done what I need to do by the time I hit ground, it wasn't worth doing in the first place.
 
9:53 PM
why the hell does that thing take 8-10 seconds to save a 30k screenshot to usb?
 
@PlasmaHH Good point I haven't checked 3 probes yet. What's the issue you're facing? So far I have a good feeling about (distributor) service.
 
@jippie one of the probes is loosing compensation when shaking... guess the trim is a bit loose. I think I will compensate and then glue down or so.
 
@PlasmaHH Why not call the distributor? Sounds like a DOA(ish) to me.
 
@jippie I dont feel like sending everything back just over that...
 
You never know, they may send you a new probe if you explain the problem
time for bed for me
night
 
10:04 PM
@jippie might try... doubt it though... its germany ;)
 
ordered mine from czech ..
 
@ScottSeidman rise time isn't that bad
but sub 10ns is a bit of a pain for a lot of things
 
@jippie cheaper or your country?
 
Can anyone explain this to me:
https://s3.amazonaws.com/s3.awesomebox.net/Kegerator+Flow+Meters/voltageVsReadings.png
The readings is what I read with my Java app through the mic port
 
@Mike capacitive coupling.
I thought we had that yesterday already? or was that someone else?
 
10:16 PM
Yea I knew that.
But I was wondering why the falling voltage causes a negative spike ?
*falling
 
because it goes down... the coupling tries to drag things to 0V but all signal changes get added properly and then go to 0V
 
I see
I am still not quite sure what "capacitive coupling" is. But it seems like it results in the change of voltage and pulls down to 0.
@PlasmaHH Why does it take so long to pull back down to 0V? Is there a way to make it faster?
 
@Mike its a function of capacity voltage and resistance. without modifying the sound system of your phone it is likely not possible
 
Hmm, I have only tested on my PC, not on the tablet... I wonder if it will be better
Is the victim's value always set to the difference or is it relative?
Meaning, if the victim is in the process of dragging down to 0v and is at -1v and the source increases in voltage by +5v, will the victim go to +5v or +4v?
 
it will likely not be -1 but mayb -0.8 so it gets 4.2 or so and then goes to 0 again
 
10:30 PM
Hmm, OK.
@PlasmaHH To be clear, if the source increases to 2v, the victim will rise to 2v, then fall back to 0v. After the victim falls back to 0v, if the source rises to 5v (increase of 3v) the victim will rise to 3v then fall back to 0v. Right?
 
10:44 PM
I thought of a way to check the code as well. If I see a downward spike and I determine that it came from flow meter 1, I can make sure that flow meter 1 is in the "ON" state as I have just detected that it has switched to the "OFF" state. If it was already in the "OFF" state then it is impossible for it to cause a downward spike and I know something went wrong.
 
11:11 PM
@W5VO Is this electronics-y enough for you guys?
0
Q: I'm having trouble finding electrical components

Corbett FrenchI was going to make a stir plate to grow yeast for brewing beer. It involves using some type of housing (cigar box, wooden box), wiring some DC voltage to a switch, to dimmer controller, to a PC fan inside the box. On the center of the fan, you glue magnets. When the magnets rotate, that rotat...

 
11:36 PM
So I got this data and I think it outlines my problem:
-- states --

0 = neither active
1 = flow meter 1 active
2 = flow meter 2 active
3 = both active

-- state voltages --

0: 5.13
1: 3.49
2: 2.15
3: 1.57

-- change in voltage --

0 -> 1 : -1.64
0 -> 2 : -2.98
0 -> 3 : -3.56
1 -> 2 : -1.34
1 -> 3 : -1.92
2 -> 3 : -0.58

(same in opposite direction)
3 -> 0 : +3.56
3 -> 1 : +1.92
3 -> 2 : +0.58
2 -> 0 : +2.98
2 -> 1 : +1.34
1 -> 0 : +1.64
When I see a peak it is difficult to tell what has just happened because some of the change of states have a very similar change in voltage.
For example, if I saw an edge that was a voltage change of -1.5, that is right in between 1-> 2 and 0 -> 1. Meaning, I wouldn't know if:
A. flow meter 1 was on, and now flow meter 1 is off and flow meter 2 is on (1 -> 2).
- OR -
B. No flow meters were on and now flow meter 1 is on ( 0 -> 1).
Well, actually. If I know the state of the flow meters I could determine what happened. For example, If I know that flow meter 1 was already on, case B is impossible and it must be case A.
In the same way, if I know that flow meter 2 is already on, case A is impossible and it must be case B.
time to write some code :)
 

« first day (1616 days earlier)      last day (3327 days later) »