So I made an oscillator! Now what..?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'll try take another set with volume definitely at 0 when I get back (I'm outta town at the mo).
I suggest just measuring resistance from each cathode to ground, with an ohmmeter, with the amp powered off, and power supply caps fully discharged.

My guess is that both triodes are at +3.5V because they both have the same-value cathode resistor - and it's much bigger than the 1.5k it's supposed to be. It isn't that hard to mistake a reddish-orange band for an orangey-red band, and so wind up with 15k instead of 1.5k.

If you actually do measure 1.5k (+/- 5% or whatever), then we do have to worry about more exotic possibilities, such as the one PRR suggested.

But if you measure something much greater than 1.5k (which I suspect you will), that is a simple matter to fix.


-Gnobuddy
 
I've attached an audio file of what the amp is doing.

In the recording, I set the volume, then alter the tone.

For some reason, the MP3 format isn't allowed (on an audio forum..) so please rename the .txt to .mp3.

Cheers.
 

Attachments

  • 20181030_9013osc.txt
    120.2 KB · Views: 25
or there is a missing ground wire, or a bad solder joint.

I think you've nailed it..

Just looking for back at the photos and schematic, I seem to be missing the ground runner between the base of the 1k5 resistors and the input / vol ground bus.
 

Attachments

  • Screenshot_20181102-121123_Gallery.jpg
    Screenshot_20181102-121123_Gallery.jpg
    341.1 KB · Views: 155
  • Screenshot_20181102-121506_Drive.jpg
    Screenshot_20181102-121506_Drive.jpg
    243.8 KB · Views: 166
Last edited:
O frabjous day! Callooh! Callay!

Looks like an easy fix, and let's hope that's the only issue! :)


-Gnobuddy

Thanks Gnobuddy,

I finally got home from interstate and was able to get time to solder in the ground line.

Noise wise, the amp is fairly quiet (50hz hum comes in at about 75% volume) and the amp sounds fantastic, even on an old Magnavox speaker pushing an acoustic youtube clip of Hotel California via crocodile clips :D
 
50hz hum comes in at about 75% volume
I used to use the 100 Hz hum to tune my guitar (3rd fret G on the 6th string) when I was a starving student, too poor to buy a guitar tuner. It isn't exact, but is within 2% of proper A 440 concert pitch. :)

Then I moved to North America, and the mains hum not only sounded all wrong (60 Hz instead of 50), but my old guitar tuning method no longer worked. :mad:

I eventually realized I could tune the 2nd fret B on the 5th string to 120 Hz AC hum, but it took me a while to figure that out! (Also not exact, but close enough to be usable until you want to play along with the CD.)

the amp sounds fantastic, even on an old Magnavox speaker pushing an acoustic youtube clip of Hotel California via crocodile clips :D
Excellent, happy days!

Is it my imagination, or has Hotel California now replaced Stairway To Heaven as the song most likely to be badly played by multiple beginner guitarists at a jam? :hypno1:


-Gnobuddy
 
In NA, pick up the wired telephone. The higher of the two tones is 440Hz.
https://en.wikipedia.org/wiki/Dial_tone
Now, that information would have come in very handy circa 1990 or so!

I'm now spoiled, and use a clip-on Snark to tune everything but my 5-string basses. The Snark lets me tune that awful guitar 2nd string B nearly 14 cents too sharp, as dictated by equal temperament.

I always found it impossible to tune that B string to my satisfaction without a tuner...whatever you do, some chord or the other will sound horridly out of tune. The Snark splits the difference, so every chord sounds a little bit out of tune, but none of them sounds altogether awful.

-Gnobuddy
 
Managed to nail the hum with one change in wiring (DC line from rectifier wasn't far away enough from lamp line).

I found there to be some ringing on a 1khz square wave. I've tamed it a bit with a 220pF across the feedback resistor.

My questions are:

how much ringing is tolerable? Do I keep chasing it to completely eliminate or is a little ok?

Is dampening the ringing via a parallel cap over the NFB resistor the 'right way' of dealing with this?

I've attached a before and after photos (unfortunately I didn't have photos at the same scale but I had checked at equal scales while selecting values and the ringing is reduced).
 

Attachments

  • 20181119_094044.jpg
    20181119_094044.jpg
    152.5 KB · Views: 75
  • 20181119_095927.jpg
    20181119_095927.jpg
    153.1 KB · Views: 63
Is dampening the ringing via a parallel cap over the NFB resistor the 'right way' of dealing with this?
Ringing is caused by inadequate phase margin ( Stability Criteria - (<b>Gain</b> Margin and <b>Phase</b> Margin) (2.010) ).

So there are essentially two ways in which to reduce ringing: either you increase the phase margin at the frequency of the ringing, or you reduce the open-loop gain of the system at that frequency, so that there is adequate phase margin at the new unity-gain frequency.

The first of these is rarely achievable - if the troublesome phase shifts are being caused by inherent properties of components in the chain (triode input capacitance, transformer leakage inductance, etc), there isn't much you can do.

So the usual solution is to take the second path, and reduce open-loop HF gain sufficiently to get rid of the ringing.

This is usually done by adding additional capacitance or resistance somewhere in the forward path (not feedback path) of the signal.

For example, you might wire a small capacitance across grid and cathode of a triode stage, or increase the value of an existing grid-stopper. Or you can wire a small capacitance from the anode to ground of a pentode, or a small capacitance with a series resistance in between the two anodes of a push-pull output stage (sometimes called a conjunctive filter, and described as follows in an ancient RCA manual: )

RCA said:
"A corrective filter can be used to correct the frequency characteristic of an output stage using a beam power tube or pentode when inverse feedback is not applicable. The filter consists of a resistor and a capacitor placed in series across the primary of the output tranformer. Connected in this way, the filter is in parallel with the plate load impedance reflected from the voice-coil by the output transformer. The magnitude of this reflected impedance increases with increasing frequency in the middle and upper audio range. The impedance of the filter, however, decreases with increasing frequency. It follows that by use of the proper values for the resistance and the capacitance in the filter, the effective load impedance on the output tubes can be made practically constant for all the frequencies in the middle and upper audio range.
The resistance to be used in the filter for a push-pull stage is 1.3 times the recommended plate-to-plate load resistance; or, for a single-tube stage, 1.3 times the recommended plate load resistance. The capacitance in the filter should have a value such that the voltage gain of the output stage at a freqeuncy of 1000 cycles or higher is the same as the voltage gain at 400 cycles.
A method of determining the proper value of capacitnace for the filter is to make two measurements of the output voltage across the primary of the output tranformer; first, when a 400-cycle signal is applied to the input, and second, when a 1000-cycle signal of the same voltage as the 400-cycle signal is applied to the input. The correct value of capacitor is one which gives equal output voltage for the two signal inputs. In practice, this value is usually found to be on the order of 0.05uF."

There is a weird issue called "pole-splitting" that occurs if you try to use a small capacitor from anode to ground of a triode to limit HF response, which is why I think this particular approach is better used with a pentode. With a triode, just increase the grid stopper, or add grid-to-cathode capacitance.

In a nutshell: do something to reduce open-loop HF gain of the amp.

Note that adding additional capacitance across the negative feedback resistor, as you did, usually makes ringing worse, not better!


-Gnobuddy
 
I usually find increasing VAS capacitor slightly stops oscillation/ringing.
That is the perfect solution for contemporary solid-state amps which have a topology similar to an op-amp, with almost all the voltage gain occurring in one single stage. :up:

But this thread is about a much more primitive design - an old single-ended Princeton 5F2A guitar amp ( https://el34world.com/Hoffman/files/Hoffman_5F2A.pdf ).

In the 5F2A, there is voltage gain in every stage, there is AC coupling everywhere (introducing LF phase shifts), and there is even an output transformer with its several phase shifts included in the feedback loop. The only reason the thing doesn't oscillate continuously is because very little NFB is used, and one gain stage is left out of the NFB loop.

I think a conjunctive filter across the output transformer primary might do the trick here. That, or a small cap from control grid of the 6V6 to ground, and an increase in R10 to a sane value instead of the useless 1.5k people insist on using.

I suggest increasing R10 to 15k (fifteen thousand ohms), then wiring a 220pF cap from G1 of the 6V6 to ground.


-Gnobuddy
 

PRR

Member
Joined 2003
Paid Member
Your "ringing" is well above 10KHz.

This has NO significance in a 80Hz-5KHz speaker.

The OT is sure to ring at some high frequency. This design has been doing this since 1957.

However--- such ringiness can *also* just be ground-ringing in your test leads. Supersonic 'scope observations need careful setup detailing.

The 5F2a has just-barely any NFB. Enuff to damp speaker bass resonance "some", not really enuff to get itself in trouble.

One thing would make me investigate further: if the ringing changes when a hand is *carefully* waved near the open chassis, especially if it grows large easily.

Don't look at the amp. Play it.
 
I think a conjunctive filter across the output transformer primary might do the trick here. That, or a small cap from control grid of the 6V6 to ground, and an increase in R10 to a sane value instead of the useless 1.5k people insist on using.

I suggest increasing R10 to 15k (fifteen thousand ohms), then wiring a 220pF cap from G1 of the 6V6 to ground.

Thanks for the suggestions and all the information on your last post. I've almost got my head around phase margins to the point where I can plot them in LT but still lots to learn!
 
Your "ringing" is well above 10KHz.

This has NO significance in a 80Hz-5KHz speaker.

The 5F2a has just-barely any NFB. Enuff to damp speaker bass resonance "some", not really enuff to get itself in trouble.

Sounds like a benign condition then. I had read somewhere that ringing can cause damage to the output tubes.. perhaps they meant oscillations..?

Don't look at the amp. Play it.

I intend to.. I just like to run through a few basic tests before I release it off the workbench.

One thing would make me investigate further: if the ringing changes when a hand is *carefully* waved near the open chassis, especially if it grows large easily.

Thanks for the tip!
 
Sounds like a benign condition then.
Mebbe so. Three out of three Fender valve guitar amps I have owned have had marginal HF stability. (Superchamp XD, Blues Junior, '65 Princeton Reverb Reissue.)

Leonidas was not known for spending any more than he had to. If tinkering with lead dress got it to stop howling so you could sell it, that was good enough. (My Blues Junior would turn into an RF transmitter unless the ribbon cables running to the output valve sockets were dressed just-so.)

That's businessman ethics. What about engineer ethics? Most good engineers I've known hate to release a product with obvious, easily-fixed imperfections. It's just part of the engineering mindset, which not only wants the thing to work, but also for it to have elegance.

So, fix the ringing or not? Entirely up to you. Are you more businessman, or more engineer? :)


-Gnobuddy
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.