How do I find the right transistor?

I have been playing around with Joule Thief's for the last few weeks. I am very interested in getting the supply voltage as low as possible. I can get light (not much) from around .2v using an ac141k germanium transistor. It's rated pretty high though. Has a pretty narrow range I can adjust with the pot (50k) I use to control the base. Same thing happens when I use high value silicon transistors in most joule thief circuits I have made.

So I need a few low value germanium transistors. A germanium equivalent of a bc107 or 2n3904 would be a great start.


But is there some way I can search for this without viewing hundreds of data sheets. That often don't have the low thresholds (ie most of the information I want is not on data sheets) on them....


Surely someone much smarter than me has been frustrated by this and made a chart or search tool or something?

The catalog for my local tech store has a good chart, that shows all the transistors in a logical order. But they don't stock Germanium or low value silicon (had to mail order bc107).

I've been staring at data sheets for three days. My brain is melting! Meeellllllttttinnnnngggg!

Can anyone offer some advice?

 

I can get tons of them on ebay.... I just need to work out which ones I want. Some can be a bit pricy.. Like I said the Data sheets aren't that helpful. Tells me what will cook 'em. But not what they need to run. Some show losses, which is a little helpful. I can geustimate from what the overload is, and get in the ball park.

Problem is there is no organization. If I was getting transistors from my local store. I can just look up or down the chart as I want higher or lower values. Check the ratings, and make an educated guess. normally gets me pretty close.

Anyone got a link to an old germanium chart or something? I don't need to know exactly the right bits. playing around and working that out is the fun bit. I just need to get in the ball park...

Like I said. looking up endless data sheets is making my brain melt...

 

Well after 3 days of trying not to ask stupid questions on forums.

I finally caved and asked my question.

Only to find this a few min later. http://www.datasheetarchive.com/dl/Datasheets-X/DSA842000-332.pdf

Facepalm


Anyway problem solved. That was quick.

I think I may hang around for a bit anyhow. been looking for a decent group of more experienced techies to answer the odd question.. I may even be able to help someone else...

 

Anyone out there interested in getting the manufacturers, some of the datasheet archives, and maybe some of the bigger retailers to all co-operate on an open source project like this?

So we can go to a website and enter in the criteria we need, and get it narrowed down to a few hundred options rather than hundreds of thousands of data sheets, with only one or two lines that actually have important info...

I've lost track of the times this week I have downloaded a data sheet. only to scroll down to page 3 and find I just wasted 5 min. Seriously probably wasted 20 hours this week. No wonder it cost a fortune to get a circuit designed. your paying an engeneer top rate to google stuff for hours on end....

Sounds like it would make things easier for everyone once it's done. As well as improving sales.

It's a bit like rule 35. If I can't find it I must create it.

Only I don't have more than a few hours a week to donate to the project...

Anyone else interested in throwing a few hours a week at this?

And you want it in less that 4 hours searching! And you certainly don't want to have to buy and try 50 different parts. Although this is probably why the manufacturers haven't done this already... But How much does it cost the industry in total labor hours?

This is not the only forum I have asked this question on. three days now and still not a decent, acceptable answer... I thought you electronics people where smart? Why do you all put up with this?

No wonder it's so hard for nOOb's to get started....

 

For my current project. I want something that will run on the lowest possible volts. while giving me a bit of frequency adjustment from 30hz up to around a khz, maybe a little higher.... I don't mind swapping transistors to get different hertz ranges.

So listing the burnout specs is pretty useless....

Like (*steve*) says, if your relying on the limits of some specification, your probably doing something wrong.

I'ts not critical now. During R&D stage. But I want to enable the third world (and everyone else) to be able to have lighting that is virtually free after a small set up cost. Making it simple enough that people not familiar with electronics can play around with it too.

Right now I am at the start of what I hope to be an open source project that will make people all over the world less dependent on centralized power networks.

My big problem finding Semis, is the specs I want are just not listed. Guessing is just amateurish, fun, But amateurish and time consuming. Why can't I just type what I want into a search and get most of the junk eliminated?

How much money and time would this save big engineering firms?

If the manufacturers and retailers want me to buy their kit. I need to be able to find it!

Datasheet archive and the likes are not very good unless you already know what part you want specs for...


All I want to do is search by specs, not part no... Doesn't seem like a huge ask. I bet the manufacturers all have this info on a database somewhere. By not sharing the info they cost everyone else in the industry, an uncountable figure.

It's like no one want's to sell transistors....


All I'm doing is asking for help. On public forums that claim to exist to help people. Yet people keep trying to shut me down? Why is this?

Are john and I the only ones that want to be able to find a part to do a job in less than a few hours?

Seriously. 3 days searching and I still don't even know if I'm in the ball park. Germanium Transistors are too expensive to buy tons of them....

I shouldn't have too!

What happened to customer service?

I go to a website looking to spend some cash. and come away downloading a LOIC and studying how to use it... Is that the sort of reaction you would want from people that could be your bread and butter?

 

So Since I asked for help selecting a germanium Transistor. with low base voltage. (under .2v is ideal, lower the better) with a slow frequency 50-1000herzt range.
If there is something non germanium that will work i'd be keen to try that too.


Do you think you might actually be able to help me with my first search?

I don't care If they don't sell it. I can find a part once I know what it is.

I don't know how perfect I need until I have some idea of whats available...

What do I type into what page? I looked at the links provided. They don't seem to help.... I asked for help with transistors, not mofsets,

I'm sure once I've done it once I'll be fine. I just don't get why it takes so many posts on 3 forums. and still no actual help....

I must say this forum has been the most helpful. But there is more of a don't ask questions attitude than actually offering help.

I have been googling for well over 20 hours. more like 40. Clearly I am missing something.

How about clearing that up rather than being rude?

Sounds like there are a few people reading this that can help. yet 12 posts and a few links later... Nothing actually useful or productive.

And this is the forum with nice people....

 

Obviously I'm new to semis. and trying to work with stuff older than me. so I don't have anything I've used before....

 

http://www.digikey.com.au/scripts/dksearch/dksus.dll?vendor=0&keywords=germanium+transistor+100hz

What am I doing wrong?

 

I got This search with some results. But I'm just not seeing the data I need to know.

How do I exclude stuff that needs more than .2v to activate the base?

Or see what its made from, ie silicon germanium hybrid etc?

http://www.digikey.com.au/product-s...stors-bjt-single/1376376?k=.2v npn transistor

 

Just Frustrated.

I'm sorry if it seems like I'm taking my frustrations out on people here.

It's just that I have been looking into the whole circuit for weeks now, and the transistor is the hold up. It doesn't need to be perfect. closeish will be fine for early prototyping. I'll have a much better idea what I want after that.

You guys are the nicest electronics forum regulars I have come across. Other forums seem to be full of intellectual elitists.

Like I said. this is the third one and the first to post a link to a site as useful as digikey.

If I do post my full idea. I think it will be here and youtube at the same time. With a website and a few other places as well.

I'd really like to get a few of the different sections of the end circuit working better than what I have, before I do that though. If it works I want lots of people replicating it quickly. So common parts (ie not germanium) would be great. just not sure if it's plausible.
That it's so hard to find the info I want is encouraging in a way. Means not many have tried the same thing....

I'd love to post a diagram or schematic. But I'm still playing around a lot. it's changing daily I'm also worried someone will patent my idea before I can open source it properly.
Although I guess if I talk about it here that may help protect it as public property. I'm really not looking to get rich off this. But others might.

The basic part I am working on at the moment, and feel I need low voltage transistors for. Is a self regulating oscillator, or joule thief. I want to take trace amounts of power and pulse it into something that can light an LED or similar highly efficient light. I'm getting there...
This is just stage one. There is more later. I might start a thread soon to help maintain my idea. Its not a new one. I've wanted to do this for 20 years. When I first came up with it, almost everyone, especially my teachers at school, told me I would have to break the laws of thermodynamics. I don't believe that is the case. I had it from a very reliable source that it was plausible. As I always knew. The more I have researched and observed, the more sure I get it will work. I think it's even been done before.....

But the first stage of the circuit needs to be able to collect low voltages at small micro amps, Then pulse it out in a more useable level.

I have my first germanium version on you-tube here
While it works well. the transistor is all wrong. I ordered that one for something else, It's just all I had. Still better than the BC347 or 2n3904 or the other common BC3-- NPN's I have tried. I have some BC107's in the post. Not expecting much from them though. I want to go for lower volts and less amps. The transistor seems to be the problem...

I'm pretty certain the base voltage requirement and the switching speed of the first transistor, or Q1 if you like, are going to be main factors. Just about everything else can be worked around....


As you will see if you watch some of my videos. I have tons of reading to do on brain melting topics like Harmonics and induction coils, and will still need help. you will also see I have a decent ground to build knowledge on...
It's not about money, it's not about fame. I just want this to be for everyone. to make the world that tiny bit better.


The digikey link answered my original question. SO perhaps it is time for a new thread. I know it's time to read my girl a bedtime story...

I'll be back a little later with more info, and probably a new thread. I'm still very new around here, so if the projects section is the wrong place to host an open source project. Let me know.

Thank's to everyone for their tolerance to my frustration and vagueness. All will become clear soon.