that might be why the 1K resistors are getting warm when they couldn't of gotten warm off just 11V by themselfs.
i touched both collectors of both transistors once (metal pad on bottom) and got a bit of a shock... had moisture on my hands because its very hot and humid where i live... Ouch.. but not too bad.. Probably wouldn't of noticed with dry non-sweaty hands.
Even with the AC turned on all day full blast it's still roasting in my room. and nope i dont live in the middle of the sahara desert
Even with the AC turned on all day full blast it's still roasting in my room. and nope i dont live in the middle of the sahara desert
Last edited:
Yes, that sounds absolutely typical for this kind of set up. Try just touching the tip of an insulated screwdriver on the collector and see if you can see an arc 😉
One real problem could be that the transistors are going to be stressed out by this spiky high voltage and possibly be prone to fail. It depends just how high the voltage is.
It might not be too high because it only shocks if i push my fingers on both collectors.. but if im just lightly touching them with dry non-sweaty fingers i barely notice anything.. could they be ok still?
I swapped both transistors around to see if one or the other is different and it seems to work a WHOLE lot better now. Even powering a 23W light bulb that it couldn't before! Fully lit and manages to turn itself on
I also added more resistors for a total of 500ohms paralleled (couldn't find the rest of my 1K's so had to use a few 10K's paralleled..)
The 20w bulb needed to be "touched" for some reason to get turned on.. weird behavior.
but it seems fully capable of starting and running a 20W AC induction motor at full speed.
I attempted to power higher power things like a computer screen just for giggles and nothing really happens but the "buzzing" turning into "whining" from the transformer... apparently it keeps trying but just can't quite do it. I see the LED on the front turn on.. monitor flashes on then off and the cycle repeats.
but the top transistor still seems to be the warmer one since i switched them (the one in its place gets warmer now.
I think it might definitely be an imbalance in the transformer.
I swapped both transistors around to see if one or the other is different and it seems to work a WHOLE lot better now. Even powering a 23W light bulb that it couldn't before! Fully lit and manages to turn itself on
I also added more resistors for a total of 500ohms paralleled (couldn't find the rest of my 1K's so had to use a few 10K's paralleled..)
The 20w bulb needed to be "touched" for some reason to get turned on.. weird behavior.
but it seems fully capable of starting and running a 20W AC induction motor at full speed.
I attempted to power higher power things like a computer screen just for giggles and nothing really happens but the "buzzing" turning into "whining" from the transformer... apparently it keeps trying but just can't quite do it. I see the LED on the front turn on.. monitor flashes on then off and the cycle repeats.
but the top transistor still seems to be the warmer one since i switched them (the one in its place gets warmer now.
I think it might definitely be an imbalance in the transformer.
The voltage across the full primary between the collectors is double that of the supply. The higher frequency harmonics of the square wave, and of the overshoot, are likely felt a little easier than a 60 Hz sine wave. And if your fundamental is high (say several hundred Hz), you'll feel that too. They say 400 Hz aircraft power is about the worst shock you can get. I've never tried that, but I've gotten the hell shocked out of me hooking up a mid/high cab hot - being fed off a Crown XS1200.
Fluorescent lights, especially the old bi-pin base types often require being "touched" to come on - even on a regular ballast on mains power. A large grounded piece of metal nearby usually eliminates this need - which is why you didn't see plastic shop light fixtures.
Fluorescent lights, especially the old bi-pin base types often require being "touched" to come on - even on a regular ballast on mains power. A large grounded piece of metal nearby usually eliminates this need - which is why you didn't see plastic shop light fixtures.
Do you consider yourself immune to shock? You actually aren't a voltmeter when your only indication is shock/no shock or dead/not dead. Seriously, this is not safe to do or discuss here and if you are restricted to only messing with low voltage, as you suggest in your previous threads, then any "tingle" is off limits (30V).
Get yourself a cheap multimeter. The old meter kind will be best when you have no need for high precision but need to get an impression of what is going on. You can also use these with inexpensive AA cells and this one even load-tests small cells e.g. New de 960TR Range AC DC Pointer Type Analog Meter Multimeter Voltmeter Tester | eBay
i want to increase the AC frequency to a higher frequency.. at the moment its kind of low. about 3/4 normal frequency.. i want slightly higher than 60hz frequency so that it stays above 60hz on load.
can I use some capacitors of certain values to get the frequency right?
and if so what value would you recommend?
also what would i need to change the placement and value of the resistors to?
can I use some capacitors of certain values to get the frequency right?
and if so what value would you recommend?
also what would i need to change the placement and value of the resistors to?
Adding capacitors wold only serve to slow it down. Decreasing the resistor value would speed it up, but eventually your base currents will be too high. Not to mention the resistors are already getting hot.
You get much better control of base current and base drive voltage if you take the base drive off a feedback winding. You can adjust the turns ratio and resistor values so you never drive the bases into reverse breakdown, yet get plenty of forward bias. Self oscillation frequency will tend to be higher as well. The winding can also be biased (from a resistor divider) to speed things up further.
You get much better control of base current and base drive voltage if you take the base drive off a feedback winding. You can adjust the turns ratio and resistor values so you never drive the bases into reverse breakdown, yet get plenty of forward bias. Self oscillation frequency will tend to be higher as well. The winding can also be biased (from a resistor divider) to speed things up further.
i cant adjust the windings. is there one of the other ways to work?
I tried using capacitors from another diagram (only had 4.7uF's) with a 1K and it seems to be working but i can't be sure from listening to the transformer...
can you describe some ideas just in case?
I tried using capacitors from another diagram (only had 4.7uF's) with a 1K and it seems to be working but i can't be sure from listening to the transformer...
can you describe some ideas just in case?
Trying to determine oscillation frequency by listening? That's easy. Feed it into another transformer to get it back down to 12 volts and hook up a SPEAKER.
I can hear the frequency from the transformer and it seems too low... Flourecent light flickers a bit.. (more watts used > more flicker) frequency is maybe 3/4 or 1/2 60hz.
the light on the surge strip is flickering a bit too.. frequency is way too low.
is there any way to increase the frequency using feedback from any other place?
using capacitors does increase frequency but ruins the actual amount of power possible on the output and somehow the 50v capacitors get warm from only 11.7v? i put them in the right way
the light on the surge strip is flickering a bit too.. frequency is way too low.
is there any way to increase the frequency using feedback from any other place?
using capacitors does increase frequency but ruins the actual amount of power possible on the output and somehow the 50v capacitors get warm from only 11.7v? i put them in the right way
Last edited:
Even if you put capacitors in the right way, they will be reverse biased part of the time. And they will always pass significant AC current. In applications like this you really need MKP's.
Unless you've got another winding on there you're stuck taking feedback directly from collector to opposite base. And the 1:1 turns ratio on the 'primary' is rather fixed.
Unless you've got another winding on there you're stuck taking feedback directly from collector to opposite base. And the 1:1 turns ratio on the 'primary' is rather fixed.
Without having the ability to test this (oscilloscope etc) you can not really take this much further. It all become guesswork.
Swapping the transistors and seeing how the circuit changes its behaviour shows that its all dependent on the individual characteristics of the parts and so not a repeatable design if you were to try and make another one.
Another major influence is the power supply itself. If you are using batteries (or AC power) then you need some pretty substantial decoupling across the supply. Some big caps like 2200uF fitted right across the supply and as near to the circuit as possible.
That might not be suitable if its a battery backup 'supercap' type of thing as they tend to have quite high internal resistance.
I was only going to use the 2F capacitor to reduce noise and provide protection for starting a 20W load that requires more watts on startup than running
and i was thinking some cheap dollar store 6V lantern batteries would be a good emergency choice (two in series for 12V)
to provide some power to keep a phone alive during a power outage or such.
and provide power to a fluorescent light for a while for us to get things around
not like we're going to be able to run an air conditioner but it should be a good cheap emergency power source?
I tested and it even lights up a incandescent bulb about halfway. rated 29 watts!
and i was thinking some cheap dollar store 6V lantern batteries would be a good emergency choice (two in series for 12V)
to provide some power to keep a phone alive during a power outage or such.
and provide power to a fluorescent light for a while for us to get things around
not like we're going to be able to run an air conditioner but it should be a good cheap emergency power source?
I tested and it even lights up a incandescent bulb about halfway. rated 29 watts!
I just found out it can run at lower voltages IF I detach one of the transistors base's from the resistor.
but the frequency becomes very high at that point.
my stereo becomes functional when in that state at lower voltages but massively less power available than if at a normal 12-ish volts
it seems to work down to 6v
and in that higher frequency state one of the transistors gets warmer than usual.
and the power light on the power strip is more sensitive to flucuations in power usage. but brighter with no load than with the transistor jumper wire normally.
but the frequency becomes very high at that point.
my stereo becomes functional when in that state at lower voltages but massively less power available than if at a normal 12-ish volts
it seems to work down to 6v
and in that higher frequency state one of the transistors gets warmer than usual.
and the power light on the power strip is more sensitive to flucuations in power usage. but brighter with no load than with the transistor jumper wire normally.
In normal use the transistors get warm but not too hot. even without a heatsink
the resistors are warmer than the transistors but still fine.. they get cooler with a heavier load. whereas the transistors get warmer with more load
I was thinking a small PC fan blowing across the transistors and resistors would be sufficient to cool them for now? till I get a heatsink for the transistors and proper resistor values? (i was thinking 100ohms 10w resistors for reliability)
the resistors are warmer than the transistors but still fine.. they get cooler with a heavier load. whereas the transistors get warmer with more load
I was thinking a small PC fan blowing across the transistors and resistors would be sufficient to cool them for now? till I get a heatsink for the transistors and proper resistor values? (i was thinking 100ohms 10w resistors for reliability)
Well yes... but all the heat generation defeats the object of making something efficient to run off batteries. Ideally you want something that would be efficient and that would maintain a set output voltage over a wide range of input voltage, a bit like an SMPS.
- Status
- Not open for further replies.