Phase Shift Oscillator

A friend (an old friend in both senses of the word) recently told me I must be "pining for tin".

Best regards, Spehro Pefhany

Gack! I remember Pine as well ;-)

What's terrible is that I used to be a whiz with batch files... now I need to get out the book plus look at old stuff I did :-) ...Jim Thompson

Here is one that i actually saw working. I helped assemble the kit a little bit.

Hey where did my link go? Trying again:

We'll see if this one works right.

josephkk expounded in news: snipped-for-privacy@4ax.com:

Very interesting. I've assembled my opamp version and apart from some issues with the pot I purchased (needs replacing), it is otherwise working. But..

I do like the idea of using a single transistor for the oscillator. Judging from a quick look your link, it is perhaps a matter of choosing the right range of RC values. I suspect that I went too big on C because my R values were too low.

Warren

josephkk expounded in news: snipped-for-privacy@4ax.com:

...

It's always interesting to look at someone else's work. I usually learn something that way. :)

Two things I found interesting about Figure 5 in:

1. The entire sub-circuit, including the phase shift oscillator is running from 60 volts.
2. I personally found the biasing arrangement of Q1/Q2, Q3/Q4 rather interesting. This exact same arrangement is used almost everywhere in this project. I'll start a separate thread for that.

Looking at Fig 5 does not indicate the supply voltage. So when I was trying to calculate the voltages and currents (while on the train), the circuit strongly suggested a 30+ supply voltage. Going to the part 2 article proved the assumption correct. It uses the same power source as the power amp stage, at +60 volts.

The magnitude of the supply voltage makes a big difference. I had originally tried to do this within 12 volts. Even when I gave up on that, I did wonder later if using the full +/- 12v (24 volts) would have been better. I should have stuck with it longer and tried that (lazy me).

You can make it work withing 12 volts, but I had great difficulty in getting any kind of usable range. It may still be possible but it appears difficult. 24 volts made it considerably easier. I can get about 3 - 9+ Hz range, which seems to be the limit at the moment. I'd like to increase that to 12 or 13Hz max.

There is one thing I can do if I can get away with just one pot for the R in RC, and use the other pot (ganged) to control the overall gain.

What I've posted below comes close. What I want to try next is to see if I can increase the gain when the R7 is reduced small. When this happens more gain is required from the stage to sustain oscillation. I use R5 to reduce the stage gain to just enough for operation. This keeps the distortion level down at the low end. Higher frequencies need the added gain and hence show less distortion and work closer to the stage's gain limit.

R5 needs to go near zero when control R7 goes below say 7K to get into higher frequencies. The current R5=47ohms works down to about R7=7K, but needs to go near zero if I drop R7 lower than that.

The problem is that the pot R7 is a 100k pot (ganged) in this case. I need to have a 2nd linked pot that goes from about 300 ohms to zero (ideally). Putting a 100k pot in parallel with a

I then thought maybe a JFET could be used in parallel with R5, but how low does the resistance of a fully on JFET go (TBD)? Probably not much below 47 ohms (I'd move R5 to the gnd side and put C4 above it, if this were to become practical).

The LTspice file for the 2n3904 version of my phase shift oscillator is pasted here:

Version 4 SHEET 1 2796 680 WIRE 1968 -144 1840 -144 WIRE 2176 -144 1968 -144 WIRE 2608 -144 2176 -144 WIRE 1840 -80 1840 -144 WIRE 1968 -80 1968 -144 WIRE 2176 0 2176 -144 WIRE 2608 0 2608 -144 WIRE 1968 48 1968 0 WIRE 1968 48 1472 48 WIRE 2112 48 1968 48 WIRE 1968 128 1968 48 WIRE 1472 176 1472 48 WIRE 1488 176 1472 176 WIRE 1584 176 1552 176 WIRE 1616 176 1584 176 WIRE 1712 176 1680 176 WIRE 1744 176 1712 176 WIRE 1840 176 1840 0 WIRE 1840 176 1808 176 WIRE 1904 176 1840 176 WIRE 1968 256 1968 224 WIRE 2064 256 1968 256 WIRE 2176 256 2176 96 WIRE 2208 256 2176 256 WIRE 2320 256 2288 256 WIRE 2336 256 2320 256 WIRE 2448 256 2416 256 WIRE 2512 256 2448 256 WIRE 2064 272 2064 256 WIRE 1584 336 1584 176 WIRE 1712 336 1712 176 WIRE 1840 336 1840 176 WIRE 1968 352 1968 256 WIRE 2176 352 2176 256 WIRE 2320 368 2320 256 WIRE 2448 368 2448 256 WIRE 2064 384 2064 352 WIRE 1584 480 1584 416 WIRE 1712 480 1712 416 WIRE 1712 480 1584 480 WIRE 1840 480 1840 416 WIRE 1840 480 1712 480 WIRE 1968 480 1968 432 WIRE 1968 480 1840 480 WIRE 2064 480 2064 448 WIRE 2064 480 1968 480 WIRE 2176 480 2176 432 WIRE 2176 480 2064 480 WIRE 2320 480 2320 432 WIRE 2320 480 2176 480 WIRE 2448 480 2448 432 WIRE 2448 480 2320 480 WIRE 2608 480 2608 80 WIRE 2608 480 2448 480 WIRE 1968 528 1968 480 FLAG 1968 528 0 FLAG 2512 256 Out IOPIN 2512 256 Out SYMBOL Misc\\battery 2608 -16 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 24 SYMBOL npn 1904 128 R0 SYMATTR InstName Q1 SYMATTR Value 2N3904 SYMBOL res 1952 336 R0 SYMATTR InstName R1 SYMATTR Value 3k SYMBOL res 1824 -96 R0 SYMATTR InstName R3 SYMATTR Value 330k SYMBOL res 1824 320 R0 SYMATTR InstName R4 SYMATTR Value 68k SYMBOL res 1568 320 R0 SYMATTR InstName R6 SYMATTR Value 100k SYMBOL res 1696 320 R0 SYMATTR InstName R7 SYMATTR Value 100k SYMBOL cap 1488 192 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 -5 33 VBottom 0 SYMATTR InstName C1 SYMATTR Value .47uF SYMBOL cap 1616 192 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 0 32 VBottom 0 SYMATTR InstName C2 SYMATTR Value 0.47uF SYMBOL cap 1744 192 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 0 32 VBottom 0 SYMATTR InstName C3 SYMATTR Value .47uF SYMBOL cap 2048 384 R0 SYMATTR InstName C4 SYMATTR Value 100uF SYMBOL npn 2112 0 R0 SYMATTR InstName Q2 SYMATTR Value 2N3904 SYMBOL res 2160 336 R0 SYMATTR InstName R8 SYMATTR Value 7.5k SYMBOL res 2192 272 R270 WINDOW 0 32 56 VTop 0 WINDOW 3 0 56 VBottom 0 SYMATTR InstName R9 SYMATTR Value 1.5k SYMBOL cap 2304 368 R0 SYMATTR InstName C5 SYMATTR Value 10uF SYMBOL res 2048 256 R0 SYMATTR InstName R5 SYMATTR Value 47 SYMBOL res 1952 -96 R0 SYMATTR InstName R2 SYMATTR Value 10k SYMBOL res 2320 272 R270 WINDOW 0 32 56 VTop 0 WINDOW 3 0 56 VBottom 0 SYMATTR InstName R10 SYMATTR Value 1.5k SYMBOL cap 2432 368 R0 SYMATTR InstName C6 SYMATTR Value 10uF TEXT 1432 528 Left 0 !.tran 0 40 0 .0001 TEXT 1488 -176 Left 0 ;Phase Shift Oscillator TEXT 2544 520 Left 0 ;April 4, 2011 TEXT 2552 -184 Left 0 ;VE3WWG

Warren

Not too bad. The EF may improve both the harmonics and frequency range. = I think i might consider a ganged dual pot for the frequency control.

josephkk expounded in news: snipped-for-privacy@4ax.com:

..

I do use two 100K resistors as a dual/ganged pot (see the "Bias Calculations" thread where I posted the revised LTspice circuit). I've since updated that again slightly when I ordered the parts from mouser. Some parts were better priced by going with slightly different values (like using 220uF instead of 200uF).

Warren