|
There is not much to say about this circuit that we have not already covered. From the scope shots it seems the base drive is rather soft, due either to a minimum of coupling between the coils or Miller effect just not having enough current drive to effectively pump the base from the resultant L-C oscillator.
Can you completely define what task you expect the circuit to perform, and why.
Some time ago I built a similar inductively decoupled circuit using an FET, but it snapped on a lot harder than your transistor version. It required a higher voltage for startup as expected and a gate biasing arrangement.
Personally, I never liked the name Joule Thief, preferring the original term which is "blocking oscillator". From watching the videos by the guy that invented the term, with all due respect to him, he never really addresses the real engineering aspects of designing such a circuit, such as optimizing base drive, and how and why the turns ratio and core material type is important.
Many variants of the humble transistor blocking oscillator have been used in many millions of consumer products over the last 50 years but it's origin is from the vacuum tube era. You will find them in small fluorescent tube emergency or camping lights, camera flash assemblies, HV boost converters etc. It is ubiquitous in it's many incarnations, all tailored specifically to the task at hand.
The desired end application of the blocking oscillator will determine it's ultimate design and no single version is best. No one type fits all applications.
If the intention is merely to steal a little of what is left from a nearly dead single cell to light an LED, a well designed blocking oscillator will work just fine, but you could also just stack a few of your nearly dead cells in series to light the led (the no component approach). By the time the cells are that dead, the internal resistance is fairly high and will nicely do the current limiting.
Well designed blocking oscillators have been used in Tektronix vintage scopes, the circuits of which are a good education in the art.
So best to define what you want to do (the job at hand) then design the circuit to fit this and still operate as efficiently and economically as possible.
Good engineering considers all this and balances it against the cost of perhaps some extra components, which spread over many millions of produced units can be considerable, or may not be necessary, depending on the market you are competing in.
If you are shooting for the most efficient design for a given application, and the cost is not an issue, you have a wide choice of topologies and components.
Regards
ION
« Last Edit: 2016-04-05, 15:49:15 by ION »
---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
|
feature is available which provides a place all members can chat, either publicly or privately.
Author
Topic: The Cool Joule is back ! (Read 4827 times)
