OverUnity Research

I need to try and build a sinewave oscillator that will work from a 100V supply and give me a large amplitude sinewave.

I did not want to use LC networks in case external magnetic fields affect the L component.

Stability is not an issue but harmonic content maybe.

I have found a great article that explains the design steps involved.
Operating frequency will be 35kHz

http://home.earthlink.net/~doncox/wec/Oscillators.html

Here's a picture of the circuit

Working through the steps one by one i hope to build the circuit.

Step 2

QuoteSelect a transistor. Almost any NPN bipolar transistor with decent gain will work. The 2N2222 is one of the most commonly available for small-signal circuits, and has a current gain of at least a hundred for currents around 150mA.

My Transistor of choice is the ZTX857

Technical/Catalog Information   ZTX857
Vendor   Zetex Inc
Category   Discrete Semiconductor Products
Transistor Type   NPN
Voltage - Rated   300V
Current Rating   5A
Package / Case   E-Line
Packaging   Bulk
Lead Free Status   Lead Free
RoHS Status   RoHS Compliant
Other Names    ZTX857
http://www.diodes.com/zetex/?ztx=3.0/3-3-2b@rid~9

Hfe of 100-300

Step 3

QuoteSelect power supply voltage. Supply voltage must be enough to turn the transistor on (0.7V) with overhead to work with (low voltage reduces your amplifier gains). Too high a voltage and we risk overheating components unnecessarily. Anything in the 5-15V range is the most common. For our example, we will use 10V.

I will use 25V

Step 4

QuoteSelect your high-end power limits. This would be due to your collector-emitter current. Most discrete small-signal components are typically capable of dissipating 100-250mW. We will choose 50mW in the transistor. It is a good idea to allow a factor of two or more in power to reduce the chance of melting or burning if the power dissipation conditions are less than ideal. (You may also find yourself with an upper limit on your power supply current which will force you to go back and adjust your values. These trade-offs are the bread-and-butter of much of practical design work.)

I will go for 50mW dissipation limit.

Step 5

QuoteCalculate power, voltage, and current in transistor output to stay within our power limits. Imagine a subcircuit which is just Rc, Re, and the transistor's collector-emitter side. The highest power dissipated in any one place is likely at the output of the transistor. Power transfer from our collector and emitter resistors (our "source") will be maximized when their sum is the same as that of the transistor (our "load"). (We are ignoring the loading effect of the first RC stage right now--remember, this is just to get a reasonable power dissipation.) The following relationships are reasonably true:

   Rce = Rc + Re
   Vce = VRc + VRe
   Pce = PRc + PRe
   IRc = IQce = IRe

Since we know the transistor gets half the power, it is like giving it half the supply voltage for the same current

   P = I * E
   I = PQ1 / Vce
   I = 50mW / 5V = 10mA

I = 50mW / 12.5V = 4mA

Step 6

QuoteCalculate lowest Rc and Re based on maximum power allowed. Let's narrow our selection problem with Rc and Re. Having an emitter resistor stabilizes a common emitter transistor circuit, but must not hog too much of the available output voltage. Today, let us choose Rc = Re. This makes a very stable circuit, increasing the sensitivity to base bias from around +/-10% to about +/-50% and reducing our dependency on a specific transistor gain, but we have just limited our DC gain hFE to 1 (=Rc/Re). Later, we will steal back our lost gain in the AC domain with an emitter bypass capacitor.

Since we know current is 10mA, power for each is half of transistor's quiescent dissipation, and voltage dropped across each is half of Vce, we have more than enough information:

   Rc = Re = VRe / I = 2.5V / 10mA = 250 ohms each (249 ohms is a standard 1% value)
   Ic(max) = Vcc / (Rc + Re) = 10V / (250 + 250 ohms) = 20mA

25mW is well within the 100mW of a typical 1% metal film resistor that is so common nowadays.

Rc = Re = VRe / I = 6.25V / 4mA = 1K5 ohms each (1K5 is a standard value)

Ic(max) = Vcc / (Rc + Re) = 25V / (1k5 + 1k5 ohms) = 8.33mA

He has this statement

Quote25mW is well within the 100mW of a typical 1% metal film resistor that is so common nowadays.

Pe = VRe / I = 6.25V * 10mA = 250mW dissipation so i am ok to use film resistors which are 600mW

Step 7

QuoteAdjust RC input impedence to take best advantage of available transistor signal output. Too low an impedence and voltage is dragged too low, too high and available power is thrown away. This circuit needs all the voltage it can get, so we'll allow throwing away some power to maximize signal voltage. At the least, we won't go lower than the 3dB point (equal impedences for maximum power transfer). So, we will keep the RC impedence > 250 ohms to guarantee modest attenuation. (Note most of the change in voltage across R will occur when XC is within a factor of several times that of R; many first order solutions then can be arrived at by assuming R = XC.) For this first-cut we will select R >= 250 ohms, which will make RC impedence roughly twice our lower limit. (There is some discussion concerning loading effects in the sections "Collector Capacitor Loading Effects" and "Emitter Capacitor Loading Effects" below.)

Step 8

QuoteSelect frequency of oscillation. For this example, we will choose 1kHz. In practice, circuit dynamics will cause frequency to be roughly twice as much. I describe the effect of the transistor stage on this in the next section after the first-order analysis. Since noise is being amplified and filtered, frequency lower limit is established by gain allowed by bypass capacitor value, and upper limit by phase shifts available in RC's and transistor. High frequencies will pass will negatively feed back through the base virtually unimpeded to ground, while very low frequencies won't find enough phase shift to reinforce themselves.

I have chosen 35kHz

Step 9

QuoteSelect C for the RC's. Typically, capacitors come in much more limited values than resistors. There are series which only have two or three values per decade, and other series will generally include these values. The most common values in use, and which embody small size versus low noise, are the simple decade values of 0.1uF and 0.01uF. We will choose 0.1uF.

I will try 2n2 or 0.0022uf for an initial trial. Not sure what the rules or limits are for choosing this.


Step 10

QuoteCalculate R to complement C at oscillation frequency. There is no trivial way to estimate the oscillation frequency better than within a factor of two or so short of using the derived equations. A 3-section RC uses f = 1 / (2 * pi * SQRT(6) * R * C), but we have a four section RC, so we use the following: *

   f = 1 / (2 * pi * SQRT(10 / 7) * R * C) (for a 4-section RC)
   R = 1 / (2 * pi * SQRT(10 / 7) * f * C)
   R = 1 / (2 * pi * SQRT(10 / 7) * 1kHz * 0.1uF)
   R = 1332 ohms (1.33k is the closest 1% standard resistor value.)

1.33k ohms is comfortably above our 250 ohm lower limit while still not throwing away too much power

R = 1 / (2 * pi * SQRT(10 / 7) * 35kHz * 0.0022uF)

R = 1.729K 1K8 is the nearest preferred value giving me 33.625 kHz

given the resistors have a 1% tolerance i can selectively choose them to get closer on the frequency as the calculated value is on the top boundary of 1% plus i will have the capacitor tolerance to play with as well.

Step 11

QuoteSelect DC operating point of transistor. When the transistor is off, all the voltage is across it, and when it is fully on, Vce is about 0.3V. So, the collector voltage (Ve + Vce) can swing from a low of 4.85 + 0.3V at saturation to a high of 10V when off. For maximum usage of this range, quiescent operating point of the collector should be:

   Vqt = ((Vmax - Vmin) / 2) + Vmin = ((10V - 4.85V) / 2) + 4.85V = 7.4V

Vqt = ((Vmax - Vmin) / 2) + Vmin = ((25V - 12.35V) / 2) +12.35V = 18.675V

Step12

QuoteThe transistor amplifier stage must be biased into its active range, so some base current must be provided by Rb. The R of the final RC stage acts with it as a voltage divider, providing a reference voltage for the base of several times the needed base current. As Vbe approaches 0.7V, the transistor begins to turn on. Most of the voltage drop across Re will come from the collector current = Ib * Current Gain of the transistor, pulling the emitter voltage up to about 0.7V below that of the base where it begins to pinch off and reach a stable state. Thus, the emitter current is effectively controlled by Re and the base quiescent voltage.

First, we know we have 10mA quiescent current across 250 ohms for Rb. While a good first guess of current gain for an unknown transistor would be 100, if we use the classic 2N2222A NPN transistor, the hFE gain is typically rated 100 to 300 (precise gain would have to be measured directly.) A first-order estimate of base current required to control our quiescent emitter current (using hFE = 200) is

   Ib = Ic / current gain = 10mA / 200 = 50µA

Base voltage divider current for a stiff reference voltage would be typically about 10 times base current:

   Ib(divider) = Ib * 10 = 50µA * 10 = 0.5mA

With the emitter quiescent DC voltage needing to be 2.5V, the base quiescent voltage is 0.7V higher

   Vb = 2.5V + 0.7V = 3.2V

The trouble is we already have the R of the final RC stage setting our base divider requirements with 1.33k ohms. This is over 20 times lower than we could get away with, but we go with it for now. We can ignore the load which is only about 1% of our source:

   Ib(divider) = Vb / RRC = 3.2V / 1.33k ohms = 2.41 mA
   VRb = Vcc - Vb = 10V - 3.2V = 6.8V
   Rb = VRb / Ib(divider) = 6.8V / 2.41 mA = 2.82k ohms

using hFE = 200

Ib = Ic / current gain = 8.33mA / 200 = 41.65µA

Base voltage divider current for a stiff reference voltage would be typically about 10 times base current:

   Ib(divider) = Ib * 10 = 41.65µA * 10 = 0.4165mA

With the emitter quiescent DC voltage needing to be 25V, the base quiescent voltage is 0.7V higher

   Vb = 6.25V + 0.7V = 6.95V I think i have it correct that i want 6.25V emitter quiescent DC voltage

Ib(divider) = Vb / RRC = 6.95V / 1.8k ohms = 3.861mA

   VRb = Vcc - Vb = 25V - 6.95V = 18.05V
   Rb = VRb / Ib(divider) = 18.05V / 3.861 mA = 4.674K

Step 13

QuoteEstimate equivalent resistance through the base to ground. From the above calculations of quiescent base voltage and current, we have:

   Rb-e = Eb / Ib = 3.2V / 50µA = 64k ohms

  Rb-e = Eb / Ib = 6.95V / 41.65µA = 166K ohms

Step 14

QuoteCalculate an emitter bypass capacitance which allows AC gain at our oscillating frequency to be high enough to sustain oscillation. The emitter bypass capacitor allows us to establish an AC operating point which is a bit different from the DC operating point of the transistor. Re creates a great deal of transistor gain degeneration (AC gain hfe here is Rc / XCe). If Ce is added in parallel with Re and is suitably large, the AC signal will find a low-reactance path on the emitter through the capacitor rather than the resistor and will return to us most of our lost gain (for our AC signal, not DC). If we want most of our gain back (say 100), we need a 100 times smaller reactance for the bypass capacitor than Re:

   XCe = Re / 100 = 249 / 100 = 2.49 ohms
   XCe = 1 / (2 * pi * f * C)
   Ce = 1 / (2 * pi * f * XCe)
   Ce = 1 / (2 * pi * 1kHz * (2.49 ohms * 0.01)) = 64µF

XCe = Re / 100 = 1K5  / 100 = 15 ohms
   XCe = 1 / (2 * pi * f * C)
   Ce = 1 / (2 * pi * f * XCe)
   Ce = 1 / (2 * pi * 35kHz * (15 ohms * 2n2)) = 137µF

Step 15

QuoteEstimate base reactance to the AC signal.

   Xbe = XC * hfe = 2.49 * 200 = 500 ohms

Xbe = XC * hfe = 15 * 200 = 3K ohms

Reading further on he goes on about the operating frequency being twice the calculated values GRRRR, i will need to build it and see.

Well the simulation does not seem to work with my calculated values, i guess i better try his calculated values to see if his works, maybe i made a mistake.
Strange why his diagram does not even use any of his calculated values. I wonder whats going on.

How much current does it need to drive Peter?

Not really sure right now because i have not designed the device to be driven, low mA would do it, the plan was to build a buffer stage anyway to match this.

There seem to be other people using all sorts of values/ratio's for biasing & Rc Re

The values i have calculated above do not seem to be in the ratio that others use for instance he said keep Rc & Re the same yet on the attached diagram there 1/10 ratio

I don't think you want Re and Rc the same; you would have no gain then.

In the first circuit the ratio is 2:1.

Another approach to this requirement, would be to use a high voltage operational amplifier, then use my "cheap and simple oscillators" document to build what you need. No guarantee it would work, but I can't see why it wouldn't at the moment.

.99

I am not even sure 100V op amps are available, could be wrong. I guess i could build one out of transistors LOL

I know i can get this working should not be too hard really.

The gain is achieved by using the emitter bypass capacitor.

It would appear because the bypass capacitors reactance is a lot less than the emitter resistor's value we get our gain back.

OK, with the bypass cap yes. I didn't see that.

HV op-amps exist, that's why I suggested it. ;)

.99

You could do it easily with a couple of 12AX7's and a split-rail +/- 300VDC supply  O0

:D

-----

I'll recant the above statement. I just remembered the output voltage was still very low. It'll never put out +/- 100V.

Maybe a garden variety low voltage circuit driving a class AB amplifier?

Quote from: WaveWatcher on 2012.02.19, 16:19:14
You could do it easily with a couple of 12AX7's and a split-rail +/- 300VDC supply  O0

:D

-----

I'll recant the above statement. I just remembered the output voltage was still very low. It'll never put out +/- 100V.

Maybe a garden variety low voltage circuit driving a class AB amplifier?

You know i have been very tempted to go down this route but i know even less about valves than i do transistor theory LOL
I guess i could easily read up, but then there's the grid transformer and the 300V rail to build, it just seems to get more complicated, and my next stage is to build a sine 0-360 Degree phase shifter as well which i have not looked at regarding valves but i am pretty sure i can do it using 3 jfet's.

I do also have 5 of the valves SM mentioned cannot remember the valve code 12b??? dual triodes.

I could also just use low voltage circuits and add an amp stage on the end to get my HV i guess but i could not really see any reason why i shouldn't built it all to work at 100V apart from working out how to convert existing circuits to work at higher voltages.

I am jumping the gun a little because i must get the sine going first but here's the phase shifter circuit

Darren
Sorry i never realized you can get op amps that high a voltage, i will look to see what i can find.

I am wondering if i could drag a 32kHz crystal up to 35kHz, but also not sure what the max working voltage of a crystal is, just looked at a datasheet and it does not say. Probably would not like 100V across it

It would certainly give me a stable sine.

Peter, I think you should not abandon this oscillator just yet, there's a lot to be learned.  I'm surprised the manual indicates the frequency could be off by a factor of 2, so much for precise engineering!  Lol

Here's my suggestions.

1). First get the amplifier biased at the mid range of the voltage (50 volts).  Make the gain a bit lower by including a resistor in series with the emitter cap. Gain is -RC/Re , and Re will be this series resistance.  Chose a gain of 10 to begin with.

2) Then, design and test the RC shift network separately by itself and make sure it produces a -180 deg phase shift at the frequency you want.

3) when you put it all together notice that you will have two 10 k resistors in parallel at the base, one from the shift network and one from the biasing on the transistor, so remove one of the resistors, or the frequency will deviate, also account for the other resistor connected to V+, in the design.   anyway some tweaking will be required on the resistor value, maybe use a pot.

4) personally, I would choose a lower voltage for the oscillator, and the high voltage in the final driving stage.

5) most importantly, have fun!  :)

EM

Hi EM
I will try what you say thanks.

I should also be able to do a bit more reading on the theory design yet as there is quiet a bit written about this type of oscillator.

PS i have it working in LT spice, not sure how good that sim is with regard to having a working circuit though, anyway i think i have enough info to get some bits to build it.

Here's a simulation of just the phase shift network.

It's a 3 section RC network, and if you place it in the feedback path of the amplifier, from the collector to the base,  it will oscillate at the frequency where the phase is 180 deg,   because the amplifier is inverting, or 180 out of phase, so the network adds another 180 deg phase and you get 360 deg phase, or zero phase, so the signal can now build up in strength, but only if the gain is sufficient or greater then one, which is the other criteria for oscillation.

On the diagram I highlighted the 180 deg phase spot, then we look at the gain of the RC network and notice that the gain is -30 dB, but this is voltage level plotted in dB, so the attenuation is about 0.032,  so we need a gain of at least 32 from the amplifier to compensate, not 10 like I said before, sorry about that, that was just a number I pulled out of the hat.  Maybe to really get it to oscillate right away make the gain 50.

The values of the capacitor and resistor can be adjusted from what I have shown, and it's desirable to increase the resistance and make the capacitance smaller by the same factor, so 2K resistance and 1 nF capacitance will keep the same frequency,  or 20k and 0.1 nF will be another pair of values.


I might build two of these myself, and offset them by 60 Hz,    lol      :-X

EM

I am glad you have taken interest, many chips around these days and it's very easy to bypass a simple cheap transistor circuit.
Do you feel the stability of the circuit will be good enough to maintain a 60Hz difference between the 2 osc circuits.

3 other things i picked up along the way

The bits can be soldered into a small blob and epoxy glued together to form a little 3 leg encapsulated osc, the guy painted it with green paint also  ;D

4 RC networks means you need less gain on the transistor

and it can be beneficial to swap the R & C over, this way the cap is to ground, i understand this can be better for harmonic rejection.

Late last night i did a bit of research into the jfet phase shifter circuit and found it's virtually impossible to find a jfet that works over 50V so it now looks like i will be running at a lower voltage and then use a final stage to shift up to HV.

The calculations above work out on the 4 stage RC simulation using 3n3 and 1k for each stage i get 180 degrees phase shift at the end of the chain with an input of 35kHz, dividing the input level against the output level it would appear i need a gain of at least 23.94

I Re did the calcs above for a 25V supply and built the osc using 5% resistors and 20% caps and a BC547C

The results are pretty poor, it runs fine but precision parts are needed to guarantee the 180 degree phase shift to get a clean sinewave.

You can see from the waveform on my scope the sine has a steeper rise because of the mismatch RC values causing an imprecise phase shift through the 4 RC networks.

Because the frequency is also unpredictable not what you calculate i will move on to another type of oscillator, maybe wein.

The main problem also is that you cannot just use pots in the RC chain to make a variable frequency because C also needs to be changed with each change of R to maintain the 180 degree phase shift.

The calcs for biasing and gain seem to work as above.

It's been very interesting going through the above process and i learned quiet a bit.  O0

Green trace is collector, Yellow trace is base.

OK this is looking good for my next try, it has amplitude stabilization and frequency control.

http://www.redcircuits.com/Page92.htm

For some reason this does not function in LTspice or Multisim  :'( i don't know enough about the sim to understand why.

Looks like i will need to breadboard it first.

OK i have this circuit working good in both ltspice and multisim now.

Anyone know how you simulate a pot in ltspice?

Built the circuit on breadboard tonight, got it to work well, there was no easy way of adjusting the frequency apart from dropping caps in, the trimmer only sets the distortion level, you adjust just after the sine stops and then back a smidgen, so i was ready to give up until i realized the circuit works well up to 30v without component change with a low current being drawn under 1ma @ 30v but the consequence of an adjustable voltage rail is that it adjusts the trimmer position which in turn alters the frequency of a stable sineus, so i managed to adjust to 35kHz no problem, when i checked out the waveform using the spectrum it was really clean, -43db harmonic distortion, so i am really happy, just need to tart it up on vero now.

I spent a bit more time on the osc tonight, i now have a good handle on the function and have added pots.

It is now adjustable between 17kHz to 70 kHz,so i have a Frequency adjust pot, threshold pot which sets the THD/Oscillation point and an Amplitude pot.

I am happy with the circuit and just need to wait for all my precision bits to arrive, i will solder the bits to the bottom of the 3 pots and use 18 turn pots glued together, then i will epoxy them up and seal in alu foil with a supply, gnd and output pin.

I can now move on and work on the 0-360 degree phase control circuit, i want 2 of these, i will feed the sine into 2 shifters so i can then control each Channel +/-180 degrees.

I build one stage of the jfet phase shifter which should give me a 0-180 variable shift.

I could not get it operating in multisim and it does not work on breadboard either right now

The circuit uses 2n3070 but these are impossible to obtain so i ended up getting some 2n5457 jfets.

On breadboard i either get 180 phase shift or 0 phase shift but distorted waveform, it suddenly flips phase at a certain point on the pot.

I am guessing the bias point must be different using the 2n5457 devices.

Looks like i need to read up and biasing or maybe try a pot in the source and see if i can find a spot where it works.

Data sheet 2n5457
http://www.futurlec.com/Transistors/2N5457.shtml

attached data sheet for 2n3070

Yes, biasing is probably causing your
unusual mode of operation.  A variable
10 MegOhm resister in the upper arm
of the biasing circuit may do the trick.

Each stage of the two stage phase shifter
operates as a phase splitter.  The signals
which appear at Source and Drain should
be equal in amplitude and opposite polarity
without any distortion.

The phase shift adjust variable resistance
portion (1.0 MegOhms) of the combiner
couples more or less of the Source signal
to be combined with the Drain signal at the
Gate of the stage following.  Zero resistance
equals 0 degrees and maximum resistance
equals 180 degrees for each of the two stages.

You'll get it going with just a little tinkering!

Be sure that the input signal (for test purposes)
doesn't overdrive the JFETs into distortion.

The type of JFET used shouldn't be critical.

Thanks for the explanation Dumped.
Very helpfull  O0

OK the reason the simulators were not working on the phase shifter was the 10nF was too large for my 35kHz frequency, 2n2 seems to work well but it would appear that a 1 meg pot is not needed more likely 50K should do it on the simulation anyway.

Just been trying to work out the coupling capacitor value between my oscillator stage and the input of the shifter from what i see i use this formulae

C = 1/(2 * pi * F * R)

If i use 35kHz as my frequency and 1meg as the input impedance of my shifter i get 4.54pF seems very low to me.

Got the phase shifter working well with 2n2 caps instead of 10nf, a 10K pot is good, as it turns out my bias is only slightly off, there is a slight amplitude difference between the sine on the source and drain so therefore when i vary i get a slight variation in amplitude, so easily fixed, thanks again Dumped your explanation of how the circuit works meant i had it working in a couple of minuets  O0

I have now refined my plan and end target design.

the whole idea of this build is to generate a sine variable in frequency, split the output into 2, then feed 2 x 0-360 degree phase shifters, so i will have 2 identical phase adjustable sinewaves, i will add a switch which can switch in a schmitt trigger circuit if needed, the output of the 2 sines/pulses will then each feed a transistor amp stage which both drive LT700 transformers, these are impedance matching transformers, these have a primary impedance of 1.2KOhm and a secondary center tapped 3.2Ohm, i am not currently sure what frequency i will be able to drive these up to but cannot find anything available that will operate at higher frequency's.

The idea being that i will be able to drive 2 transformers out of phase using either sine or pulsed to see if there are any signs of interesting effects to study, this is a sort of analogue version of my digital controller.

So next step is to get an adjustable Schmitt Trigger working, i will place a pot in place of the 1K resistor this will allow me to adjust the hysteresis and therefore the pulse width, combined with the already available 100k for setting the trigger threshold voltage.

The Schmitt seems to simulate pretty well.

Not yet sure about the final amp stage, i have found a diagram as follows, maybe over complicated for what i need, ultimately i would like to be able to drive bifilar coils instead of a speaker, i would think that as long as my bifilar impedance at the driven frequency is about 3 Ohms then i should be OK.

I have seen single transistor LT700 driver stages, are the benefits of using the below diagram going to justify the complexity of the output stage.?
Does anyone have any idea if this will work at 35kHz? I will probably need to build it and see i guess.
http://www.hoofbags.me.uk/geekiestuff.html

I built the amp in the last post seems to work, i used a 3ohm resistor as a load but have a problem i get some sort of oscillation on the output stage.

The scope shot green is across the primary of the first transformer or LT44
the yellow trace is across the secondary of the final stage transformer with a 3.4 ohm load across.

So does any one have any idea what may be causing this oscillation.

The other boards are coming on as well, still waiting for my caps from hong kong for the sine board.

I have built and tested both phase delay boards now they work pretty well, i still have a slight amplitude variation between 0 and 360 degree shift and did not get to the bottom of this, but i think this should not be much of a problem.

So the boards picture we have sine osc left, 2 phase boards in middle and amp on right.

A vero layout of the phase delay board
A vero layout of sine osc although i have not tested this yet

Quote from: Peterae on 2012.03.14, 21:16:19
So does any one have any idea what may be causing this oscillation.

That's the busker amp?

You're most likely getting a 180º phase shift at the oscillation frequency. The common bandaid for that is a series RC "snubber" network on the output to ground, usually a 1 Ohm and 0.1u or 10 Ohm and 0.1u or so. Do some research and experimentation with the values.

You can check this page for more details:
http://sound.westhost.com/amp_design.htm#op-stability

QuoteThat's the busker amp?

That's right

Cheers Darren i will have a read in the morning.
Drat i didn't want to add components on the low impedance coil because i want to drive bifilar coils without added components, although maybe they wont affect it.

Quote from: Peterae
...

So does any one have any idea what may be causing this oscillation.

Your scope shots may provide the clues:

The input signal to the Push-Pull Output Stage
seems to have some sort of parasitic oscillation
riding on it.

What's showing up in the output may simply
be this parasitic amplified.

Something to look at is the input power from
your supply - does this "oscillation" also appear
on your DC input as "ripple?"  If so, you may be
experiencing the old "motorboating" effect.

Ripple on the DC supply can feed back into the
prior stages to produce spurious oscillations.

Try loading the circuit lightly to see if that makes
any difference.

Thanks Dumped
I will look at that also, it maybe this scope, the Hantek DSO has always suffered a noisy trace on mv settings even when shorted.
It does seem strange the circuit does not use a zobel network, and must assume it works without one, although i will also look at this also.

Many amps have the network, and this one especially should also. It's a cobbled together design, not of the utmost fidelity, and stability is probably hit and miss.

Do you know of any instances of stabilizing it on the primary side of the output transformer, surely if i filter on the primary it will have the same effect, i am really trying to avoid extra components on the secondary that could interfer with the coils i am connecting to it.

Quote from: Peterae on 2012.03.15, 17:54:24
Do you know of any instances of stabilizing it on the primary side of the output transformer, surely if i filter on the primary it will have the same effect, i am really trying to avoid extra components on the secondary that could interfer with the coils i am connecting to it.

Sorry Peter, I thought that was implied. Yes, the Zobel network would go on the output of T3/T4, in parallel with the 2.2-4.7 Ohm resistor. Changing the value of that resistor may have an effect on the oscillation as well, but you're safer if you include the Zobel network.

Oh that's great news, sorry my confusion all i kept seeing was resistor capacitor from the speaker  O0
That's cheered me up a bit.

This amp is rough, i understand that but i wanted something cheap to build for a trial, i will need to switch to something more robust at some point.

I lost Patience today waiting for my caps Grrr and used caps i have in stock tonight to get my sine oscillator working, it works well, I ended up placing a 680Ohm resistor in series with the 1k threshold pot, this gives me an effective frequency range of 25kHz to 65kHz, i may extend this to just over 70kHz but am not too worried right now and added 4n7 caps for coupling each phase delay to the sine source.

I decided to leave the amp stage for a few days and get the osc system up and running, this way i can use my final circuit for testing the amp.

I am tempted to build an agc control to keep the amplitude of the various waveforms the same, otherwise i will may end up with problems matching the 2 separate waveforms.
I will be able to switch between sine or square as well and each circuit mismatches the amplitude, i also have the problem that the phase shift circuits slightly alter the sine amplitude as the phase shifts, if i do add a transistor agc circuit i can do away with the amp volume pot as well.
I will need to do tests to see how well this will work with sine and squarewave inputs and how well it controls the amplitude of each input signal.

Thanks for your help chaps, i wired the whole lot up in a box and now when i drive the amp from my sine osc and phase shifter i do not get the nasty oscillations.  O0

May be my old signal generator picking up noise or something.

Well after many hours i have failed to get the above circuits working as i want them, i've learnt a lot and have been lead in the direction that i should have followed in the first place.
I have been following Spherics approach to delaying one pulse before another pulse i realize i should be following the approach that SM took

QuoteI originally got the idea from electron circuits which use vacuum rectifiers like the 5U4 GB or 5AR4 etc.
If the plate voltage is not rectified then it is AC with a potential 60 Hz frequency.
That combines with the 5 volt 60 Hz in the coil of the htr transformer and generally amounts to nothing. In fact the power of the 5 volt transformer amounts to nothing.
It is an insignificant power supply except when the two transformers get slightly out of phase with each other, or when they are connected in reverse of one another.
Then you can measure all kinds of things going on.
You can generate all kinds of hash and multiple frequencies, and I do mean all kinds.

The next build will be to drive 2 transformers out of phase using the phase shifter, but using AC mains as the sine source, i will drive a 450-0-450 transformer from the AC mains supply and drive a 5V 3Amp transformer using the mains phase shifted frequency.

So my first and only problem with this build is to develop a Class A power amp that will drive a winding to give me 5V AC @ 3Amps.

Anyone know what the chances of success would be using a fet as a class A to drive a transformer winding, as opposed to using a transistor class A?
Are there any benefits for using a Fet instead of a Transistor.

I have chosen to drive the filament transformer using a class A because i do not want any chances of switching spikes in the waveform.

I am also hoping to upconvert the 50Hz mains to the 3rd harmonic and be able to switch this in to drive the filament phase controlled winding.

After a bit of thought about the Class A amp i wonder if anyone has ever used a variable voltage regulator, LM138 is a 5Amp device and varies between 1.2V and 32V if i drive the adjust pin using my sine wave i maybe able to drive a transformer winding.

This thread is now continued here for further testing of SM's heater valve phase tests.
http://www.overunityresearch.com/index.php?topic=1498.msg22804#msg22804