I wish he would comment in the text on why he used the base collector junction in the picture, the text only deals with the use of the base emitter junction.
He mentions the differences between BE and BC junctions when used in this type of application but I do not know why he used the BC junction here either. The breakdown voltage will be higher but that does not matter and it will have slower recovery which seems like it would be a disadvantage.
He mentions the differences between BE and BC junctions when used in this type of application but I do not know why he used the BC junction here either. The breakdown voltage will be higher but that does not matter and it will have slower recovery which seems like it would be a disadvantage.
I couldn't see those pages of the book ... but he does talk a bit more about it in this article EDN was nice enough to dig from their archives :
http://www.edn.com/electronics-blogs/designing-ideas/4311721/Bob-Pease-on-bounding-and-clamping-techniques
Thanks, that indeed had interesting tidbits in it. Pity the links to the (presumably) better images are dead, the EDN inlined jpg's were not super duper. But still readable enough.
That is common with EDN and related online trade magazines. I expect it more often than not.
Their comment system is even less reliable than the EEVBlog forum.
That is common with EDN and related online trade magazines. I expect it more often than not.
And the sad part is that the machine on which those jpg's are still exists. They just decided (or probably didn't even decide anything) to not maintain those linked images. From which you can usually extrapolate amusing characteristics about the CMS & backup solutions they use.
Their comment system is even less reliable than the EEVBlog forum.
The EEVblog forum is not reliable?
Haven't lost a single attachment or post since it's been running...
Long term reliable, with more than a few service interruptions though
Sometime the https://archive.org/web/ can help getting old stuff.
Agreed. Unfortunately linked images is not the wayback machine's strongest suit. It's rather hit & miss ... mostly miss. Case in point, the hi-res images from that EDN article where not to be found that way.
The first time I glanced at that EDN article I gave up too soon. After reading mr.Flibble's recommendation I went back and downloaded and magnified. Well well worth it! But one thing disturbed me, In the section marked " So you say you need a diode" Pease writes:
"For extremely low leakage .. consider using collector-base junctions. If, for instance, you need leakage lower than 100pA, then it's often much less costly to test transistors with a 98% yield to this spec ....
Some low leakage transistors to consider are the 2N930, 2N3707 and 2N4250. However most other small signal transistors have an excellent yield to a leakage spec of 30 pA max at 5V, even though the spec sheet guarantees 10nA."
edited for brevity and all transcription errors mine.
Now if someone dropped a bag of transistors in my lap I would have no clue as to how to characterize their leakage. Do you need a pico-Ammeter? or can it be done some sneaky cheap way with a bridge and some op amps?
Simple, just use your 10M input impedance DVM and put the transistor in series with the positive lead, and apply a voltage that is equal to Vceo of the transistor to the whole lot. Then simply bin according to the measured voltage, the lower the voltage the lower the leakage.
Their comment system is even less reliable than the EEVBlog forum.
The EEVblog forum is not reliable?
Haven't lost a single attachment or post since it's been running...
Every couple weeks I have to erase the EEVblog cookies (there is a forum thread discussing this remedy) and I have lost posts often enough that I sometimes remember to save long ones locally before pressing post just in case. Usually Firefox allows recovery of the post text but it only takes losing it once. Preview and post occasionally return an error requiring the thread to be reloaded.
Given the choice, I prefer email based forums because they are both faster and more reliable.
Update: It was unable to access the forum for hours not long after I posted this. Instead I got this:
Lost connection to MySQL server during query
Connection Problems
Sorry, SMF was unable to connect to the database. This may be caused by the server being busy. Please try again later. Connection Problems
Sorry, SMF was unable to connect to the database. This may be caused by the server being busy. Please try again later. Error establishing a database connection
Now if someone dropped a bag of transistors in my lap I would have no clue as to how to characterize their leakage. Do you need a pico-Ammeter? or can it be done some sneaky cheap way with a bridge and some op amps?
Simple, just use your 10M input impedance DVM and put the transistor in series with the positive lead, and apply a voltage that is equal to Vceo of the transistor to the whole lot. Then simply bin according to the measured voltage, the lower the voltage the lower the leakage.
This yields 10 microvolts per picoamp which is good enough for most applications but many transistors will have a leakage low enough to require either an electrometer or some type of charge measurement to be made with something like an integrator.
You also have to be careful because higher resolution voltmeters usually lack a 10 MOhm shunt on their lower voltage ranges so the shunt has to be added externally.
The integration method is preferred at the highest sensitivities because it has lower noise. It can be low enough to pick up cosmic rays if the integration capacitor has too great a volume.
An Open Source Multimeter would be an interesting project. But I wouldn't expect to see a rotary switch on it. Rather, I would expect it to use new techniques like CTMU (Charge Time Measurement Unit). Most measurements could be made by just connecting the device to a part or to points in a circuit. It could detect a component and report resistance, capacitance, inductance, beta, etc. If connected to an active circuit it could detect voltage, current, phase, frequency, duty cycle, etc. and report ... all without the user turning a single dial. Maybe the user would press a start button ... likely built into a probe. It could have an audio annunciator so the user wouldn't have to take his eyes off what he is doing. It could take voice commands.
P.S. I'm new here. Does this forum app have a threaded message option?
Tapatalk works we with eevblog forums.
Hello!
I'm considering building a multimeter as my first serious project.
Something simple at first, but I think it is had a good analog/digital mix. After searching a bit I didn't find many information, do you know about some classes, books, articles that could help me?
Best regards!
Reminds me of the TNG-Tricorder
Reminds me of the TNG-Tricorder
I see the similarity. Beam me up Scotty. Did they move actual atoms to do this or just use information and atoms that are already on the Enterprise , somewhat like a 3d printer. From a philosophical point of view it makes a big difference if you are the one that is being transported.