I am trying to create a simple schematic. My potentiometer and Transistor parts have pin numbers on them in the drawing. Is there any way I can hide these numbers?
I found this conversation: how to hide pin numbers But the work-around to change library symbols that are read-only does not make sense to me as a new user.
Attached is screenshot:
Mmm not that I know, they are actually pretty useful, but as a work around, you can change their color to the background color in v5.1.9 (an the stable branch in general I assume):
Preferences>Preferences>Eeschema>Colors>Pin number
The above example is not perfect, I just chose white color but you get the point.
I didn’t looked what there were written.
At element in library you can right-click and Export Symbol…
Save it under your name in selected by you directory - you will get new library with one element.
Using Preferences - Manage Symbol Libraries you add your new library to the list.
You can do with your new element there what you want.
Funny! I’m exactly the opposite. I insist on pin numbers for all my symbols, even non polarized ones. (This is for documentation reasons, I find it easier to refer to R3 pin2 (or R3-2 for short) instead of “the side of R3 closest to U45”…) What I did was create my own personal library parallel to the provided libraries and put copies of the standard symbols but with pin number visibility turned on. You just need to do the same thing, but have the symbols in your personal library with the pin number visibility turned off. You’ll need your own personal libraries eventually anyway as there is no reasonable way for KiCad to have all possible symbols for all possible components already provided.
Check the FAQ area (there should be a link to the FAQ in the top bar of this site just to the left of the search magnifying glass) for some well written tutorials for the latest stable version of KiCad. The first post in that FAQ area should be an index for the other posts to make it easier to find the one you want.
Potentiometers are polarized parts. You do not want to swap the outer pins, as that would make the volume go down when you turn the knob clockwise.
I’m with the others. I do not know of a method to turn of pin numbers of specific schematic symbols without changing them in the library. And, as you know, the default libraries are read-only, so that means you have to copy a default schematic symbol into a personal library and then use that modified part in your schematic.
The FAQ part is indeed the most upto date tutorial like documentation and has about 50 articles, and it is also searchabl, for example for library managment
Thank you all for your input and suggestions. Hiding the pins with a color change seems like my best option from you folks.
I would recommend to copy those few symbols into a personal library and do it the “regular way”.
It does take a bit of effort to do the library management in the beginning, but after that it’s not too difficult and it would circumvent possible future problems.
Changing text colors to set them the same as the background color may lead to problems, or not work when exporting to .svg or .pdf or for printing.
What if you post process an .svg in inkscape and change the background color, or make it transparent?
Note that you do not have the potentiometer wiper connected. Without that connection you have only a fixed resistor which is probably oversized and overpriced for its circuit functionality.
There are some transistors with different pinout. While pin numbers for a SOT23 (for example) do not guarantee anything, they might help keep you out of trouble. Having said that, I think I probably turn off pin numbers on many SOT23 transistors such as MMBT3904.
So I am very pleased with KiCad. Program works very well. As a new user “clicking all over the place” I managed to lock the program twice during my schematic drawing but no harm done.
Here is my final drawing:
And…
If you want to start learning electronics than it is good idea to do that PCB.
But if you think it will be useful then:
Thank you Piotr.
You have found a “typo” / error in my circuit. NPN device under test J1_e (emitter) should go to ground, not the positive B1 rail. Here is updated schematic:
I address your talking points in no particular order…
There are over 1 million articles on “Transistor” AND “hFE” (using Google search) so I suppose the parameter is worthy of a few entries on device datasheets from their manufactures. As you may have figured out, I am a life long student of electronics - self taught - so this week I am studying hFE. So I really value all the comments from this forum.
Vtot = 9 V
Vtot = 9 V
The concept for the circuit is to put a constant 10 uA current into NPN or PNP (not both) base of device under test. Then measure DC current gain (hfe probably not hFE) through device emitter / collector. hfe is then read from the 10 mA scale after multiplying by 100 on DMM.
I bought a cheap $11 (USD) meter from Amazon. The hFE socket did not work for me. I also bought a BK 2704C multimeter with hFE socket (the most expensive one I can find) that includes measurement conditions in the manual. I also bought a PEAK atlas DCA pro semiconductor analyzer that also reports hFE among other parameters and curves.
But why, really, should a person test hfe in their lab or workbench?
Personally I like this answer: Topic: How come transistor beta hFE test is only on cheap DMMs? (Read 23244 times)
My answer to this question would be that experienced EEs seldom need to measure it these days. I have been designing (discrete and other) circuits since the mid 1970’s and I do not remember the last time I wanted to make that measurement. Add to that…
The experienced EEs probably buy better DMMs although I would not rule out buying a cheap one.
Good circuit designs usually rely on the Hfe being something less than that specified in the datasheet. Most popular small signal transistors have low current Hfe > (50 or 100) and there is seldom a need to have more than that.
At least in the USA, good small signal transistors are likely to be < $0.10 each if you buy 100 at a time. I do not need so many different types because an MMBT3904 or MMBT4403 is very versatile. So I can use good small signal transistors in my projects with modest investment.
The area where Hfe might be more of a concern is with power bipolar transistors operating at higher current. But testing those is not within the capability of a meter powered by a typical 9V battery. Still, normal design practice usually means you design to the specification sheet and not to the parameters measured on one sample.
If I were going to make a tester as a project, I’d consider this design: https://www.mikrocontroller.net/articles/AVR_Transistortester#Introduction_.28English.29 which does LCR, diodes, and transistors. On the other hand I have already bought one of the Chinese copies of this open source project so I don’t have incentive to any more. They can supply and assemble the parts far cheaper than I can source them. The only assembly I did was the acrylic case.
It’s amazing how adding a bit of intelligence in the form of a cheap MCU can vastly increase the feature to hassle ratio.
Are you sure about that?
I buy them for the LCD with an ATMEGA tagged on the bottom.
Fun development boards, and they come with lot’s of different LCD’s. (HD44780, graphical both monochrome & color, O-LED) and all for small prices.
Incentive to build the project myself that is. Look at the photos, they are not HD44780s. And the board is custom, with a ZIF socket for the DUT.
But if you find other boards, with or without displays, good value, more (micro-)power to you. I’ve been recently playing with an ESP8266 board (with WiFi) that cost just over $2.
Ok. I understand how D1 and D2 works, but I don’t like it.
When you measure PNP as Q1 collector is opened all Q1 emitter current goes through its base to D2. But that all current is in range about (3.7-0.7)/(330k+100k) = 7uA. For me that means that you work at 5.6V diode knee rather than at constant voltage characteristic section. Fortunately 5V1…6V8 Zener diodes have sharp knee and hFE measurement don’t cares on noise in voltage at D1 or D2.
Such little current through D2 is the reason why simulation gives you 5.3V at 5.6V diode.
If you design something using that one transistor you have, than it is the only case when you can base on its actual measured parameters. But what if that transistor will fire in future and you will need to replace it?
If I design something with npn or pnp transistors I can’t rely on measured parameters. My design should work with any transistor of that type you buy. So I have to match the worst possible parameter values. That values come from datasheet. I don’t remember me measuring hFE later than when I was a teenager.
Very true. I already bought the $130 (USD) version from PEAK in England. Has a nice software interface to capture traces and control parameters from PC. I am building this circuit as a student wanting to learn KiCad, constant current circuit design in general, and a little bit more about transistor parameters as I am mostly a hands on learner.
But the Arduino/AVR micro devices seem very flexible and solve lots of problems my little circuit overlooks.
So my road towards studying hFE began recently when I built a 350 ps pulse generator from a 200 Volt DC-DC source and the avalanche breakdown behavior of a 2N2369 (where only some devices work under these conditions) - see Appendix D “Measuring Probe-Oscilloscope Response” pp.93-5, in Linear Technology 1991 High Speed Amplifier Techniques Application Note 47 by Jim Williams. Link
So the circuit is just a one transistor circuit - looks simple - but I was for warned that Mr. Williams was a great analog master with few engineers at his caliber. So naturally I wanted to build his circuit.
Dumb luck got my first attempt to work but did not want to test it on my brand new Tek MSO64B - 1GHz scope. I only tested it on a 40 MHz scope that suggested it was working but not sure. So I packaged up my circuit to do final test before moving to my new fancy scope but it now failed and is still failing.
OK, I am not a “Mr. Williams Master” as of yet. But to fix my circuit I decided to explore lots of stuff including improved soldering technique for GHz designs and trying to understand why only some 2N2369 transistors work in avalanche mode. I suspect I just miswired something when packaging up my circuit, or maybe it is my use of a 20V 2pf capacitor replacement for a bit of speaker wire micky measured at 2pf in a multi-hundred volt design. My transistor still tests ok but it only has an hFE of 50 (just inside minimum on datasheet) when my other 5 units purchased for this project are new at hFE = 200.
Anyway, back to hFE. Maybe hFE has something to do with why some transistors will avalanche and others don’t. But my hFE test circuit is a dog - I agree. I put a NPN DUT into the PNP socket and it got too hot to touch. That part probably dropped a few hFE points doing that 
. And so, I am learning about electronics at 63 when you learned as a teenager if I am reading you right.
My breadboard for this hFE measuring circuit:
Gosh! I once built an avalanche circuit around 1978. I wanted to measure cross talk and pulse propagation between two narrow and closely spaced traces on an 18 inch long test pcb. As I remember, some avalanche specified transistors were available but they were prohibitively expensive. Maybe $30 each. So I was able to get 2N2222 or similar to avalanche with maybe 100V Vce and 10 ohms base to emitter. With one avalanching transistor pulling the pulse up and another pulling it back down, I was able to create a pulse width of a 2 or 3 nSec (??). This was an interesting experiment! I determined that coupling between two close traces was dominated by inductive coupling and not capacitive coupling. But I do not remember what was the underlying reason to fund this experiment.
I am 63. When I was learning electronic - click at my avatar picture.
I have never tried to do avalanche breakdown circuit. I have never had 1GHz oscilloscope. The one I build as a student was 5MHz. And during first 5 years of our firm it was the only oscilloscope we had. Modern LVC logic gates are for me enough fast
