Sope question

[bsco]

Using an old Brunelle style scope.....(MO-1251 is another model....same scope).....I want to measure the signal on the plates of the output tubes.....the plate voltage is in the vicinity of 450V...the scope will only handle 300V dc max at the input......If I use a x10 probe, will that be safe or do I need to physically make up a resistive voltage divider........which I am not too keen on as the high plate voltages will be exposed and everything just sort of tacked together......so, am I safe with the x10 probe/ or should I build a resistive divider network and safely install it in a plastic box?????
Cheers
P.S. The reason I want to do this is there is some sort of oscillation when the customer has the master volume on 10 and the channel gain at anything over 5 in the vibrato channel......the clean channel appears to be worse......when scoping the output, once it breaks into oscillation, the sine wave vibrates up and down and then finally breaks into a maze of super-imposed square waves and sine waves.....I have managed to get the channel volume in the vibrato channel up to 8.5 before it breaks into this problem by moving the position of some of the wires..... but the clean channel is still bad........and just to add....I did find a loose ground connection from the filter cap for the bias supply.....the securing nut was not really tight so I removed it and the lug, cleaned the chassis and lug and re-installed but this make no difference.....

[Tom Phillips]

Take a look at the tread at http://music-electronics-forum.com/t31492/ for a discussion on the topic. There is also a schematic of a 100X Tektronix probe circuit that you could uses as a reference build. (You don't need the cap & inductor compensation components if you are not concerned with preserving the exact shape of the waveform)

[bsco]

Take a look at the tread at http://music-electronics-forum.com/t31492/ for a discussion on the topic. There is also a schematic of a 100X Tektronix probe circuit that you could uses as a reference build. (You don't need the cap & inductor compensation components if you are not concerned with preserving the exact shape of the waveform)

Thanks Tom......so I guess that My scope as it currently stands can't handle that level of DC on it's input....and to add to that...the max dc voltage that the probe can handle must be greater than the max DC voltage that you want to measure.......even if I opened up the scope and put a 1000Vdc rated cap at the scope input connector to both channels, the max DC that the probe can handle would then be the deciding factor of how much DC voltage one could measure safely...It looks as if I will have to build the resistive voltage divider to connect to the plate of the vacuum tube and then connect that to the scope probe.....and then by doing that, I can measure both AC and DC signals without causing any damage.......keep in mind that I am dealing with Audio signals...mainly vacuum tube amplifiers......In SS amps the voltage is nowhere near these levels....

[Tom Phillips]

Thanks Tom......so i guess that My scope as it currently stands can't handle that level of DC on it's input....

That would be my opinion. Can't safely and/or repeatedly handle for sure. Details in post #21 of the thread I referenced above. There have also been several discussions pointing out the high level voltage spikes that occur on the output tube plate pins. You need to consider the whole waveform that ends up at the scope's input.

[Chuck H]

My scope is good to 400V at the input. I've used a 10X probe on the plates or output of many amps. One tenth of 450V is 45V. With the 10X probe your scope should be fine. If your probe is like mine there is a switch to select between 1X and 10X. Just be sure the probe is set correctly to 10X before checking a plate.

EDIT: I guess I don't see how using a separate divider is technically better than changing the division of a single divider (as the link says is the significant difference between 10X and 100X probes). But it looks like I shouldn't say you'll absolutely be fine doing it. Though I can't see why not. So proceed with the 10X at your own risk. I will

[Enzo]

yes, just because the input is rated at 400v doesn't mean it will blow up at 401v. On the other hand it doesn't mean it won't either. I have abused my scope many times and got away with it.

[bsco]

yes, just because the input is rated at 400v doesn't mean it will blow up at 401v. On the other hand it doesn't mean it won't either. I have abused my scope many times and got away with it.

I would think that your scope or scopes would have higher ratings than the one that I have......I will have to check the data sheet on the probe that I have.....I can't afford to chance abusing the scope I have......it may not be much but it's mine and it works...I would like to keep it that way...I do have the schematics thanks to G1......I'll take a look at the voltage rating of the input blocking capacitor and let you know what it is.......Thanks for the reply....and hope your retirement is going well.....

[Jazz P Bass]

If the amp is oscillating, why focus on the output tube plates?

I have never scoped the plates

From what you described, the problem lies before the output section.

[bsco]

Just checked the schematics.....when the selector switch is set for DC, .the capacitor that is connected to the input to the first Fet, is .01/500V.....this cap is after the volts/div. selector......there is also a capacitor connected to the AC function of the switch that selects AC or DC coupling......this one does not have a voltage rating on the schematic....... .022uf....I will have to remove the cover to check the voltage rating......

[bsco]

If the amp is oscillating, why focus on the output tube plates?

I have never scoped the plates

From what you described, the problem lies before the output section.

Everything before the output is clean.....no signs of oscillation anywhere......even up to the input of the PI tube....and I even scoped the grids of the outputs.....also clean....that is why I wanted to check the plate output.........I can double check it again for you........the distortion is definately at the speaker output jack....
Cheers...............

[nickb]

It is important to consider the voltage rating of the probe too. A typical x10 probe will be rated around 600V max. You'll get away with it...until you don't. A x100 will be more like 2,000V.

If you do chose to make your own x10 divider you will need to pick a high voltage type resistor for the higher valued one.

[bsco]

It is important to consider the voltage rating of the probe too. A typical x10 probe will be rated around 600V max. You'll get away with it...until you don't. A x100 will be more like 2,000V.

If you do chose to make your own x10 divider you will need to pick a high voltage type resistor for the higher valued one.

Thanks for that.....
Cheers

[Chuck H]

Everything before the output is clean.....no signs of oscillation anywhere......even up to the input of the PI tube....and I even scoped the grids of the outputs.....also clean....that is why I wanted to check the plate output.........I can double check it again for you........the distortion is definately at the speaker output jack....
Cheers...............

There aren't any amplifiers or acoustic elements within that limited circuit block. The output is almost certainly part of the oscillation problem. It has to be in phase correlation with an earlier circuit, the complete picture including an amplifier, in order to oscillate. If you limit yourself to the OT secondary and anything past it (which I suppose is the speaker lead and speakers) you won't be able to solve this.

If the amp has a NFB loop the location of the lead can be important. I always put the series resistor right at the NFB source to prevent having what amounts to an unshielded speaker lead running through the amp.

I recently solved an issue with a reverb circuit that had near ultrasonic oscillation at higher gain settings. The loop was between the recovery stage grid and the amps speakers somehow. Grounding the speaker baskets solved the problem. That's not a really well and understandable solution but it illustrates my synopsis of a feedback loop. The oscillation only shows at the output of the loop. That's why your only seeing it at the speaker jack.

Is the master volume stock or is it added or the amp is a homebrew? I would first suspect lead dress or grounding for the MV circuit. Sometimes moving leads can reveal things. Sometimes not. But look for anything like long, unshielded leads on grid pins and high current leads near earlier amplifier circuits.

If the amp never did it before, but does now, look to see if the HV rail preamp decoupling nodes share any like phase amplifiers. If so you can try paralleling that filter cap with another to temporarily see if there is a feedback loop within the power supply.

One trick is to take an alligator clip lead and clip one end to the chassis. Clip the other end to a 470pf to .01uf cap and poke it into the signal chain starting with the PI. Chances are good that the earliest place in the amps signal chain that the cap stops the oscillation is near the beginning of the feedback loop.

Daisy chained grounds for like phase circuits are bad. Especially if output circuits and preamp circuits are involved. This is the sort of ground problem to look for and it would include something like a preamp stage decoupling cap being grounded with an output stage filter through a shared lead before being grounded. If it IS a ground issue it's more likely something about how the MV is implemented so be sure and scrutinize that circuits ground scheme.

HTH

[bsco]

If the amp is oscillating, why focus on the output tube plates?

I have never scoped the plates

From what you described, the problem lies before the output section.

Ok Jazz....I checked everything again.......The clean channel will produce a very ugly waveform which looks like the 400Hz signal superimposed on a AC power supply waveform... looks like the modulated signal on an AM radio.....that is when it is scoped at the speaker output across an 8 ohm dummy load....When I scope the plates of both V1 and V2, they are clean......The owner had bought some power supply caps and had installed them himself.....they are properly installed........maybe one of these could be bad with high volume.....After scoping everything again I might drain the caps and do an ESR check......there has to be something weird here.....the amp works great on a low to moderate volume......but with Volumes on crank, the distortion is unreal......and it is in both channels....no reverb or tremolo....so it only leaves the PI, Output or Power Supply...Thanks for all the info so far........
Cheers,
Bernie

P.S..... On a side note here....which has nothing to do with this amp, I picked up an old 1930's RCA console radio earlier today for $100......it works but does need some work....hoping it will be a winter project......and with any luck, this winter's project......

[Jazz P Bass]

Just a thought.

If all is as it should be at the output tube sockets (screen resistors, grid resistors, cathode grounding) & you still see something at the output that is not at either grid input, suspect the OT.

Aside: keep in mind that there is that 47pf (or close to it) capacitance of the scope lead itself.
This capacitance can actually interfere with the tests.
A one meg series resistor on the probe can help to swamp the effect.

( I was working on a signal conditioning circuit where hooking up the scope probe at the 'right spot' fixed the problem.)