Phase shift oscillator not working

[mkvl]

Hi, I made this phase shift oscillator circuit exactly like the picture below but it doesn't work, I have checked everything and changed the component values many times, I don't have an oscilloscope so I use a LED to test the circuit by connecting the base of the LED's transistor with the output of the oscillator, the result is that the LED It lights up for a very short time and then turns off, I thought that the output of the oscillator is DC and therefore the oscillator does not work, and I dont know why?!! Or what's the problem?!! I made the circuit exactly as in the diagram below, so I'm confused.
I appreciate any help resolving this problem, thanks.

[Circlotron]

The oscillator is probably running but the base-emitter junction of the transistor in series with the LED is rectifying the signal and charging up the cap that is in series with the output until the DC voltage across the cap prevents any further light from the LED. Replace the transistor with a link and put a 1N4148 or similar diode in reverse polarity across the LED and see how you go.

[mkvl]

Hi, thank for reply, you mean to remove the transistor in serie with the led and connect the led with diode on reverse with the output of oscillator, I tested it didn't work, the led didn't power up, by the way, which side of the led to connect to the output? The positive or negative.

[Circlotron]

Like this.

[mkvl]

Hi, brother it seems working I solved the damn problem, I'm happy brother because my back is crooked whit this damn circuit, it seems that the main problem was in the values of the coils or transformer, I'm not sure because I have changed the value of all components many, many and many times but the last move I did before it work I changed the coils, thank you bro for help, I just want to ask you one more question, the LED is working fine now but I just want to make sue if the output is really an AC signal?, I'm using a small capacitor of 0.1nf at the output and the output is connected to transistor base which is in series with the LED, so the output is definitely an AC signal isn't?! Thank you.

[Circlotron]

The transistor that is connected to the LED has one connection labeled "G".
What does that mean?

[Ian.M]

Hi, brother it seems working I solved the damn problem, I'm happy brother because my back is crooked whit this damn circuit, it seems that the main problem was in the values of the coils or transformer, I'm not sure because I have changed the value of all components many, many and many times but the last move I did before it work I changed the coils, thank you bro for help, I just want to ask you one more question, the LED is working fine now but I just want to make sue if the output is really an AC signal?, I'm using a small capacitor of 0.1nf at the output and the output is connected to transistor base which is in series with the LED, so the output is definitely an AC signal isn't?! Thank you.

What coils?  What transformer? 

There are none in the original oscillator schematic, or your hand drawn version with the LED, or even in your photo of destroyed circuit boards!

[srb1954]

Hi, I made this phase shift oscillator circuit exactly like the picture below but it doesn't work, I have checked everything and changed the component values many times, I don't have an oscilloscope so I use a LED to test the circuit by connecting the base of the LED's transistor with the output of the oscillator, the result is that the LED It lights up for a very short time and then turns off, I thought that the output of the oscillator is DC and therefore the oscillator does not work, and I dont know why?!! Or what's the problem?!! I made the circuit exactly as in the diagram below, so I'm confused.
I appreciate any help resolving this problem, thanks.

Did you even solder the wire connections to the various loose components?

I can't see much evidence of solder in the photo so I would regard nearly every joint as being a potential point of failure.

[mkvl]

 😂, yes you are right, I'm confused thinking because I'm having hard time with couple circuits, I was having broblems about a tuned oscillator circuit i have made so i have posted here in this site about it, but I have moved on to working on the phase shift oscillator and it didn't work, so I posted about it, and then I returned to work on the previous tuned oscillator and I solved the problems, so I get confused and I replied here about it when the post is about the phase shift oscillator.😂 
This is the circuit I'm talking about below.

[mkvl]

Letter G at transistor it just mean the ground.

[MathWizard]

I remember trying to make those years ago, and never getting it working either.

The base voltage is awfully low in the 1st circuit, it's starved for current. I changed it to 47k and 10k, and it oscillates no problem in LTspice. Also instead of the last 10k on the right going to gnd, I put a ~5800R in series back to the base, then 5800+(Rbb||rpi)=10k, to make the calculations quicker. Now I'm trying to calculate the the gain and frequency like in this pdf, on RC phase shift oscillators.

https://www.rcet.org.in/uploads/academics/rohini_91077289317.pdf

[gnuarm]

In the phase shift oscillator, does anyone think it odd that R3 and R7 are in parallel?  I would expect there to be a capacitor between R3 and R4 in the printed schematic.  This would allow R4 and R7 to set the bias point of the base, without interference from R3.  No?

To prevent the added capacitor from interfering with the feedback phase shift, I would use 100 nF or more.

Or, remove R7 and adjust the value of R4 to give a good bias point with R3.   That's probably a better solution. 

I'm also concerned that the relatively high impedance of the phase shift circuit will be loaded by the relatively low input impedance of the transistor.  The capacitor in the emitter leg, C4, lowers the input impedance of the transistor, potentially reducing the overall gain of the circuit to less than 1 and also affecting the phase shift of C3/R3. 

The output drive at the collector of the transistor can be very low impedance.  But then the phase shift network is a higher impedance.  I'm not sure it can drive the input at the base, properly.  Maybe increase C1, C2 and C3 to 10 nF and lower R1, R2 and R3 to 1 k.  R4 will need to be adjusted to maintain the bias point, or add a larger capacitor between R3 and R4.

Maybe find another example circuit for this design.

[Terry Bites]

[I think that parallel combo is odd. Even if they'd gone an order of magnitude bigger, I'd still worry. WTF?]

Not to point out the obvious again but.... the dc voltage at the base is barely enough to turn on the transistor. Calculated from the circuit values, it is 660mV, that puts the emitter at 0V and therefore the collector current near zero.
Make R4 56K (57k it's not an E value) Remove R7 (its another screw up) increase R3 to 13K. I hope that fixes it.

The schematic on the net you built this from is wrong. Circuit In digest?

[mkvl]

I struggled with it but eventually I made it work, it's all about the components values.
Thanks everyone for help.

[MathWizard]

If you had something like a binary counter/Period doubler IC, like a CD4024, so long as the collector voltage swing is big enough, 1 or 2 of those chips, that would allow you to get an output you could see on an LED, like 10Hz with ~9 stages, from ~10kHz input.

Or maybe some 555 timer circuit that reaches some threshold after say 1000cycles of the osc.

I would definitely recommend 1 of those $20 ebay/amazon mini-scopes. I didn't use my very much before getting a real scope, but being able to see your results of something like the RC osc, is way more fun. And easier.

I found the T-parameters for the amp section, now I'm still working on the feedback section. What a mess of algebra, just for a few parts, even with all the same R and C values.