I'm bored (DIY oscillator)

[saturated]

Btw sitting here doing some thinking

The circuit that worked utilized capacitors 100x smaller than what I tried 

And the resistors that I tried were orders of magnitude larger 

So going back to the original generic schematic that I could not get to work....it used the polarized capacitor symbol so hence I used electrolytic capacitors in my trial runs. 

Also referring back to the schematic for the tube generator and I presume MFD means microfarads but idk

You cannot view this attachment.

But it could be that there's no reason to compare the two (SS vs tube) because they are completely different animals

At any rate I may try using the smaller capacitors and resistors and see if it works

[saturated]

So the original generic configuration that was attempted I made a big reduction in values for resistors and capacitors

Increasing the power supply slowly I noticed there was something that rapidly appeared and disappeared as voltage continued to increase

So I went back to the narrow range where it would appear and carefully dialed it in and got this

You cannot view this attachment.

What's wierd is if I decrease voltage the frequency decreases and if I increase voltage the frequency increases 

Because it was bugging me that amplitude was not responding to voltage level 


Before I shut it down I decided to write down some voltages

The first set is taken when there was the visible waveform or apparition or more likely abberation 

Then below that is when the voltage was increased to cause the abomination  to collapse and become a straight line

You cannot view this attachment.


Hmm looking at this I don't think the base is supposed to be more positive than the collector but I'm not sure

Maybe I need to increase the Q1 base resistor

[J M Fahey]

Base is not more positive than collector, you have huge voltage drop across that resistor connected to +B
In a working circuit, base is around 0.7V above emitter which is well below collector voltage.

Horrible waveform is typical of a "relaxation oscillator"
Those were much used in TVs, Radars and such, to generate sweep signals.
Think of it like a simplified multivibrator.

Frequency is NOT stable, changes a lot with voltage.

[g1]

Also, what your meter is calling DC when there is a waveform present is not all DC.

[saturated]

Also, what your meter is calling DC when there is a waveform present is not all DC.

The voice of sanity/reality has checked in 


You cannot view this attachment.

[Kaz Kylheku]

The "butterfly-like" transistor pair arrangements, where one of them has the base and collector tied together are "current mirrors". The transistors are made together on the same substrated by the same fabrication process and have nearly identical VBE curves. And that includes temperature behavior, plus: the transistors are close together on the same chip and so thermally linked: one won't be hotter than another with a different temperature gradient of its VBE curve.

The VBEs of the current mirror pair are tied together, so whatever VBE is one transistor is the same on the other and that drives the collector current. This relationship causes the collector currents to be nearly identical (i.e. mirrored). By setting up the collector current in one half of the current mirror, the other half generates the same collector current.

The schematic shows several current mirrors tied together: Q8/Q9, Q10/Q11 and Q12/Q13.

The main purpose of current mirrors in amps is that current sources look like very high impedances.  It's like a trick for producing a very high resistor, that appears to be sitting on a high voltage. This is particularly valuable on a chip because it's hard to produce large resistors on a chip and takes up space. E.g. if we put a 1M resistor on a 10,000V supply, we would have a 10 mA current source. Provided we used the current source with loads much smaller than 1M, it would look stiff. If we make a similarly stiff 10 mA current source using transistors, it kind of looks like a 1M resistor on a 10,000V supply, in some ways. If we use that as a load resistor for an amplification stage, it produces a lot of gain. Using a current mirror (or other current source) as a load resistor is called "active loading".

You know like when you use a larger resistor for a tube plate, or BJT collector, you get more gain in the stage.  But there is a limit; the bigger resistor you use, the more you choke off the current. You can make up for that with voltage to a point. One problem with that is if the transistor shuts off, current stops flowing and then the full voltage appears on the transistor, which may not be rated for it. Active loads will never generate a voltage in excess of the supply. While within their limits for current regulation, though, they look like a giant resistor with a large voltage behind it.

These current mirrors clutter the 741 representative schematic, making it twice as "scary" as a schematic in which resistors play similar roles. Once you understand the mirror pattern, you mentally treat it as one unit.

[saturated]

Hey thanks man that was very generous of you to take the time to  offer that info I appreciate it

I need to start paying attention to impedance as it's obviously a major player in these systems

I was just reading about it here actually

You cannot view this attachment.

As G1 had me reading about it and this

You cannot view this attachment.

I did notice today that I had a pack of 741s also