 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Transistor tester/curve tracer
| Author | Message | ||||
| Alastair Senior Member  Joined: 03/04/2017 Location: AustraliaPosts: 161 |
Being older I have lots of transistors of a wide variety. I find that the actual specs of a particular device varies widely and the datasheets confirm this by having big ranges. I plan to use some in projects that are non urgent and have been looking at testers that will give me the info I need. I have a simple unit TC1 that tells me what things are - very useful, accurate and cheap. I want something that will give specs and curves. The UK Peak Electronics DCA75 Pro is the only lower end unit that does what I want and much more. It seems overkill for what I want/need. It also has limited voltage range for curve tracing. I am now considering building a standalone MM+ unit with a touch screen as I find having to connect up to a PC just to do a test is inconvenient. I have searched around to see if I am reinventing the wheel as I have often done. I would be interested in any projects that might give me ideas. I have a basic design in my head but would like other ideas as well. Plan to have a device that needs to be told what the device is, have the right connections etc and setup by touch screen. Max voltage, increments, load resistor all user set and then it will display key parameters and show curves. Memory for parameters for matching. Curve display against stored set to help matching. All power supplies in box, probably powered by a LiPo in box with usb for charging and perhaps connection to printer. Have not printed direct from a MM. Suggestions of easiest way to print curves and text without having to link to a PC? Cheers, Alastair |
||||
panky Guru  Joined: 02/10/2012 Location: AustraliaPosts: 1116 |
@Alistair, Plenty of lurkers so it clearly is of interest to 'shedders. There are quite a number of tester boards that output to an xy cro - this might be one avenue -use the testing board to control the voltages and currents which is the fiddly bit and use the Micromite to emulate the cro in xy mode as well as controlling the actual tester board. Will be watching with interest. Regards, panky ... almost all of the Maximites, the MicromMites, the MM Extremes, the ArmMites, the PicoMite and loving it! |
||||
CaptainBoing Guru Joined: 07/09/2016 Location: United KingdomPosts: 2171 |
Morning Alastair. Fashion a centronics port on the MM wiki, flow , then feed the specific commands to the printer. Grab an old Epson compatible printer (cheap enough on ebay but the supply can vary and the prices are tending to rise now). Most printers supported graphics modes, you simply send a control character set followed by a series of bytes representing the data. Its a long time since I did graphics printing direct from a processor - It was PPI and RH plots from wind/weather WF33 RADAR on old Z80 stuff back in the 80s, but the figure 1/216th of an inch comes to mind for the stepping size we could achieve in graphics mode. There are much better heads on here who will be able to shed more light on it. The results were perfectly acceptable - proper graphs on the printer (not ASCII representations) by outputting graphics data directly to the head. Seem to remember it did a reasonable job. I think the problems you'll encounter are around buffering the data and then rendering it as a series of rasters. Depending of the MM you intend to use, RAM can be a problem. If the graph data is processed in a block view, you won't need to buffer every single point in the graphing area but I reckon it will still stretch to several 10s of KB, so memory-mapping the area in RAM is likely to be unrealstic for any decent resolution. By dividing the graphing area into blocks (of say 8x8 which is convenient) you only need to explore if any trace extends into that block, if not just output 8 zero-bytes, if it does, I would render the entire 8 bytes and output them, then the next etc... then the next line and so on until you are done. This would minimise memory use as you only render a small part and not the entire graph paper. It will get trickier if you have multiple traces (which is why I memory mapped the whole thing   )my 2p |
||||
| Alastair Senior Member Joined: 03/04/2017 Location: AustraliaPosts: 161 |
@Captain Thanks for this. Years ago in the days of DOS I did a lot of stuff with dot matrix printers to print logos and bar codes. Can just remember the pain but not the detail. Perversely last year we moved into a new house and as a part of the declutter I threw away many old computers, monitors and dot matrix printers. I think I will end up using something like an HC-12 link to a PC and send the data that way. At the moment I am focused on the generation of the data but I need to have in mind how I will document the data. Don't want to build this into a major project as it started in my mind as a side issue. @panky Have done a lot of searching and reading as a distraction when I am fed up with projects I should be working on. Not all electronic. There are lots of things to do with cro display of curves. The problem is that they are mostly display only and give little quantitative data. Your idea to take the simple device output and quantify it is something I have not considered. I have been thinking more of driving the DUT (device under test) in a known way and hence the output has real meaning. Hmmmm. Currently I have been playing with a Maxim DS1846 which is a tiny I2C chip that has two 10k 100 position and one 100k 256 position pot, memory and other stuff. You set the pot values by I2C commands and can store calibration values etc. My thinking is to use this to set the bias values etc. It is 5v so I would probably use some 741 op amps (I have heaps) which can work +/-15v to actually drive the DUT. I think I can make it work on paper. This would give me enough voltage range for what I want. My challenge at the moment is how to have the full polarity flip when dealing with NPN & PNP etc. Initially I thought about having the MM floating and having the DUT connected as needed. I found this difficult and could see many opportunities for the famous smoke to escape. My current thinking is to keep it simple with the MM in a fixed supply situation and the switch the levels as needed by op amps etc. It may needed more components but I think it will be easier, for me, to design. My paper version has promise. This is a low priority project so will take some time to mature. Cheers, Alastair |
||||
| Frank N. Furter Guru Joined: 28/05/2012 Location: GermanyPosts: 979 |
Hi Alastair, take a look to this project. It's a Transistor-Tester with an ATmega... - maybe interesting for you? Frank |
||||
| Alastair Senior Member Joined: 03/04/2017 Location: AustraliaPosts: 161 |
@Frank Thanks. I have seen another derivative of the same original design. I will look at this in more detail. Cheers, Alastair |
||||
 Print this page |
 |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2025 |
