I don't have a function generator; this is just the amp's noise with the inputs shorted. The scope is uncalibrated, so I'm not sure how much to trust the signal magnitude it shows. Also, this looks really bright here, but I had the intensity down to a point where I could barely see it. My digital camera had a hard time figuring out what was going on, and these are the best pictures I have.
My question is about the spikes in these images. What do they mean? And if it's oscillation, any ideas on what it could be, and what I could do to fix it? All the images are the same signal, with different timebases - 5, 2 or 1 ms/div. And I just noticed that the spikes seem to appear with fairly consistent frequency, once every 4ms or so. Which means something kicking in with a frequency of 250Hz? 240Hz? I have a tube rectifier on the B+ supply, and all heaters are AC. What event(s) occur(s) with a frequency of 240Hz in an SE tube amp?
Also, every alternate peak of the main noise signal is of a different amplitude. That's just a result of 60Hz noise combining with 120Hz noise, right? I can see the waveform changing as I adjust the humpots, and this is about the lowest I've been able to get it.
Other images:
http://mywebpages.comcast.net/saurav/2a3/scope1.jpg
http://mywebpages.comcast.net/saurav/2a3/scope3.jpg
http://mywebpages.comcast.net/saurav/2a3/scope5.jpg
http://mywebpages.comcast.net/saurav/2a3/scope6.jpg
http://mywebpages.comcast.net/saurav/2a3/scope7.jpg
Thanks a lot in advance,
Saurav
My question is about the spikes in these images. What do they mean? And if it's oscillation, any ideas on what it could be, and what I could do to fix it? All the images are the same signal, with different timebases - 5, 2 or 1 ms/div. And I just noticed that the spikes seem to appear with fairly consistent frequency, once every 4ms or so. Which means something kicking in with a frequency of 250Hz? 240Hz? I have a tube rectifier on the B+ supply, and all heaters are AC. What event(s) occur(s) with a frequency of 240Hz in an SE tube amp?
Also, every alternate peak of the main noise signal is of a different amplitude. That's just a result of 60Hz noise combining with 120Hz noise, right? I can see the waveform changing as I adjust the humpots, and this is about the lowest I've been able to get it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Other images:
http://mywebpages.comcast.net/saurav/2a3/scope1.jpg
http://mywebpages.comcast.net/saurav/2a3/scope3.jpg
http://mywebpages.comcast.net/saurav/2a3/scope5.jpg
http://mywebpages.comcast.net/saurav/2a3/scope6.jpg
http://mywebpages.comcast.net/saurav/2a3/scope7.jpg
Thanks a lot in advance,
Saurav
The short peaks look like standard diode noise to me, check the ground path between the transformer CT, 1st filter cap and the rest of the circuit.
Are your heaters floating? That's a possible source of this kind of noise.
That it's going up and down every other cycle means there's 60Hz with 2nd harmonic distortion, i.e. 120Hz, or probably more accurately, 120Hz with 60Hz modulation.
That probably means the rectifier is unbalanced - one side of the PT's CT has a lower DCR than the other, so one charges the filter cap faster. The filtering will attenuate 120Hz at least twice as much as 60Hz, damping the 120Hz making the 60Hz appear stronger.
Please get your scope calibrated so you can take measurements with it.. and adjust the focus too
Tim
Are your heaters floating? That's a possible source of this kind of noise.
That it's going up and down every other cycle means there's 60Hz with 2nd harmonic distortion, i.e. 120Hz, or probably more accurately, 120Hz with 60Hz modulation.
That probably means the rectifier is unbalanced - one side of the PT's CT has a lower DCR than the other, so one charges the filter cap faster. The filtering will attenuate 120Hz at least twice as much as 60Hz, damping the 120Hz making the 60Hz appear stronger.Please get your scope calibrated so you can take measurements with it.. and adjust the focus too
Tim
Ahhh, focus! I hadn't even noticed that little screwdriver slot. That's the control I was missing. Thanks Getting this scope calibrated will probably cost more than what I paid for the scope. I'll certainly do it some day, but not right now.
When you say 'standard diode noise', would a tube rectifier have this noise too? I don't have any silicon diodes in this amp. I'll check the ground path as you suggested.
Well... the 2A3 heaters are grounded through the cathode bias resistor, and the 6SL7 heaters should be grounded through the voltage divider used to get the heater closer to the upper triode's cathode potential. I can try putting a cap on that voltage divider to see if it takes more AC to ground. I think I've seen that done before. And all my heaters are AC.
That's what I thought it was. I noticed that adjusting the hum pot seemed to affect mainly 60Hz noise (both on the scope and audibly), the 120Hz hum stayed mostly unaffected. So a less-than-perfectly balanced hum pot would cause this too, I think.
Thanks a lot,
Saurav
Getting this scope calibrated will probably cost more than what I paid for the scope. I'll certainly do it some day, but not right now.When you say 'standard diode noise', would a tube rectifier have this noise too? I don't have any silicon diodes in this amp. I'll check the ground path as you suggested.
Well... the 2A3 heaters are grounded through the cathode bias resistor, and the 6SL7 heaters should be grounded through the voltage divider used to get the heater closer to the upper triode's cathode potential. I can try putting a cap on that voltage divider to see if it takes more AC to ground. I think I've seen that done before. And all my heaters are AC.
That's what I thought it was. I noticed that adjusting the hum pot seemed to affect mainly 60Hz noise (both on the scope and audibly), the 120Hz hum stayed mostly unaffected. So a less-than-perfectly balanced hum pot would cause this too, I think.
Thanks a lot,
Saurav
For a cheap and crappy cal, you could take a known AC voltage, say filaments (is your voltmeter accurate?) and set horizontal to display one 60Hz cycle in 16.6msec, and for vertical remember that the voltmeter reads RMS, so multiply by 2.828 for peak-to-peak voltage.
That's the procedure recommended for my 1971-era Heathkit scope, it has a 1Vp-p output on the front panel. Now that's accurate calibration!
Tim
That's the procedure recommended for my 1971-era Heathkit scope, it has a 1Vp-p output on the front panel. Now that's accurate calibration!
Tim
Hmm, hadn't thought about that. I had thought about using a signal/function generator, but then I wouldn't know if that was calibrated.
My voltmeter is a Radio Shack digital cheapie, but it should be good enough on 60Hz sine waves, I would think. That's something to do for tomorrow.
Thanks once again,
Saurav
My voltmeter is a Radio Shack digital cheapie, but it should be good enough on 60Hz sine waves, I would think. That's something to do for tomorrow.
Thanks once again,
Saurav
Alright, I did a rough check on the calibration, and it seems to be reasonably close. I don't have a way to accurately generate signals in the millivolt range, but filament supply voltages seem to be correct both in amplitude and in frequency.
I did some more checking, and the B+ supply of the 2A3 does not have these spikes. They appear on the plate of the 2A3, and on the OPT secondaries. So I decided to check the grid of the 2A3, and this is what I got (that's the best I could do with the focus control. It gets a little thicker when I change from viewing ground to viewing the signal, and the camera's auto-exposure thickens it some more):
That's at 10mV/div. The trace is the same at the cathode of the upper 6SL7, and the plate of the lower 6SL7 (the driver is an SRPP, I removed the cathode bypass cap a few days ago to see if I liked the sound better this way). The AC ripple on the 6SL7 B+ is pretty small, to the point where my scope has a hard time triggering off it, and here's the best I could do (5mV/div):
Is this normal? That looks like a really strange waveform to me... almost triangular with other stuff in there. And doesn't triangular mean a lot of odd order harmonics?
Any ideas would be most appreciated.
Thanks,
Saurav
I did some more checking, and the B+ supply of the 2A3 does not have these spikes. They appear on the plate of the 2A3, and on the OPT secondaries. So I decided to check the grid of the 2A3, and this is what I got (that's the best I could do with the focus control. It gets a little thicker when I change from viewing ground to viewing the signal, and the camera's auto-exposure thickens it some more):
An externally hosted image should be here but it was not working when we last tested it.
That's at 10mV/div. The trace is the same at the cathode of the upper 6SL7, and the plate of the lower 6SL7 (the driver is an SRPP, I removed the cathode bypass cap a few days ago to see if I liked the sound better this way). The AC ripple on the 6SL7 B+ is pretty small, to the point where my scope has a hard time triggering off it, and here's the best I could do (5mV/div):
An externally hosted image should be here but it was not working when we last tested it.
Is this normal? That looks like a really strange waveform to me... almost triangular with other stuff in there. And doesn't triangular mean a lot of odd order harmonics?
Any ideas would be most appreciated.
Thanks,
Saurav
What time/div on horizontal?
My guess would be that last waveform is charging spikes (the filter cap only charges at the peak of the cycle), and getting in there by ground loop. This could happen with something like this:
Buss wire >---HV CT------3rd filter cap-----1st filter cap---
With the 1st filter cap's return path across the 3rd cap (which is the cap on the 6SL7's supply... this assumes a CLCRC supply, I don't know what you have ), the resistance of the buss wire, however small, produces a voltage due to the high current spikes, and these are coupled by the filter cap to the HV rail.
If you switch the position of the caps, the charging spikes stay between the CT and 1st cap, while the 3rd is isolated from that circuit.
BTW, do you have a 1x or 10x probe?
Edit:
My best guess: the noise is obviously coming from the first stage. Try putting some caps from heater winding to ground (about .1uF) and see what that does.
Tim
My guess would be that last waveform is charging spikes (the filter cap only charges at the peak of the cycle), and getting in there by ground loop. This could happen with something like this:
Buss wire >---HV CT------3rd filter cap-----1st filter cap---
With the 1st filter cap's return path across the 3rd cap (which is the cap on the 6SL7's supply... this assumes a CLCRC supply, I don't know what you have
), the resistance of the buss wire, however small, produces a voltage due to the high current spikes, and these are coupled by the filter cap to the HV rail.If you switch the position of the caps, the charging spikes stay between the CT and 1st cap, while the 3rd is isolated from that circuit.
BTW, do you have a 1x or 10x probe?
Edit:
My best guess: the noise is obviously coming from the first stage. Try putting some caps from heater winding to ground (about .1uF) and see what that does.
Tim
Hello, Suarev!
I've browsed this forum from time to time, but this is my first post.
I'd just like to offer a word of caution. From my experience, I've come across transients like these in many different types of equipment. I've spent more hours than I'd like to admit chasing these snivets down, only to find that they're not really coming from the circuit in question.
I suggest a sanity check. Start with the same conditions that produced the 'scope pictures in your first post. The ground clip from the 'scope probe should be attached to your chassis ground. Check first to see that the pulses are still there. Next, leaving the ground clip grounded, unhook the probe tip, and hold it near the measurement point, but not on it. Still see the pulses? Next, hold the tip of the 'scope probe to ground at the same place that the probe's ground clip is grounded. Now do you still see the pulses?
If it seems like the pulses are everywhere. then try removing the probe's ground clip from the chassis, and clip the ground and probe tip together. Did that make the pulses go away?
There are many possibilities that should be considered before tearing into the guts of you amplifier. Bear in mind that you are using the 'scope in unbalanced (single-ended) mode, which requires connecting 'scope ground to your chassis ground. Together with the third wire AC power connection, ground loops are created. The 'scope amp is also at a very high sensitivity, so it will amplify signals in that loop(s), irrespective of their source. It is possible that the glitches you see could be coming from the 'scope itself, of some other piece of equipment that's on nearby (TV, VCR, microwave oven, etc.). Try turning things off, one at a time, to see what effect it has.
I didn't mean to run on so long, but when I saw this thread, it was like a conditioned reflex. Or perhaps better to say deja vu?
Hope this was helpful.
M. W.
I've browsed this forum from time to time, but this is my first post.
I'd just like to offer a word of caution. From my experience, I've come across transients like these in many different types of equipment. I've spent more hours than I'd like to admit chasing these snivets down, only to find that they're not really coming from the circuit in question.
I suggest a sanity check. Start with the same conditions that produced the 'scope pictures in your first post. The ground clip from the 'scope probe should be attached to your chassis ground. Check first to see that the pulses are still there. Next, leaving the ground clip grounded, unhook the probe tip, and hold it near the measurement point, but not on it. Still see the pulses? Next, hold the tip of the 'scope probe to ground at the same place that the probe's ground clip is grounded. Now do you still see the pulses?
If it seems like the pulses are everywhere. then try removing the probe's ground clip from the chassis, and clip the ground and probe tip together. Did that make the pulses go away?
There are many possibilities that should be considered before tearing into the guts of you amplifier. Bear in mind that you are using the 'scope in unbalanced (single-ended) mode, which requires connecting 'scope ground to your chassis ground. Together with the third wire AC power connection, ground loops are created. The 'scope amp is also at a very high sensitivity, so it will amplify signals in that loop(s), irrespective of their source. It is possible that the glitches you see could be coming from the 'scope itself, of some other piece of equipment that's on nearby (TV, VCR, microwave oven, etc.). Try turning things off, one at a time, to see what effect it has.
I didn't mean to run on so long, but when I saw this thread, it was like a conditioned reflex. Or perhaps better to say deja vu?
Hope this was helpful.
M. W.
I think both those are 5ms/div.
I'll check this, but I've done that once with a preamp and ended up with a horrible buzz, so I've tried to be careful about that ever since.
Right now it's set to 1x, there's a switch that has 1x-Reference-10x positions. It's a pretty cheap probe.
Yup, I'll give that a shot.
In that case, welcome, and I'm glad that my question prompted you to respond
I already tried some of these. I see noise similar (both in shape and in amplitude) to the last scope trace between the chassis and the PS or signal grounds. And when I touch the two probe leads together, I get a nice flat line. I haven't checked for the probe tip held near the ground tip.
Yes. Both were on the chassis, pretty close to each other, and I still saw noise.
Yes, my amp and my scope both have 3-prong plugs. And for all those traces, I had the probe's ground connected to signal ground (cathode bias resistor of whatever tube's plate I was measuring). Hmm.. now that I think about it, this means that I shouldn't measure just across the OPT secondary taps, because my secondary isn't grounded. Or maybe I should ground the secondary. I couldn't see or hear a difference either way.
I was pretty disheartened to see the noise when comparing chassis ground to signal or PS grounds. I've used regular hookup wire for the grounds (and paid attention to the sequence in which I connect grounds), and I thought that meant that I'd have to re-do everything with thicker ground wire. But when you say "ground loop is outside of the amp" - does that mean that this might not be the amp's fault? That would be reassuring
Thanks,
Saurav
Calibrate? Schmalibrate!
In the trigger department of your scope, there should be a button marked "Line". It will be near buttons selecting "Ch1", "Ch2" or "Ext". "Line" triggers your scope from the AC mains, and is an excellent way of quickly determining whether noise is main derived (if it is, it will be stationary, if it's oscillation, it will roll through the screen).
As for calibrating your oscilloscope, there are lots of adjustments that can drift, and who's to say they've all drifted in the same direction? In other words, "If it isn't wrong, don't fix it." What you do need, though, is to calibrate the user! Either you need a good electronics engineer to show you what everything does, and why, or you need to read a book to tell you how to drive an oscilloscope. However, the first thing you can do is to look for small (front panel) pots with "cal" marked at one end, and make sure they are in the "cal" position. "Variable" or "fine" sensitivity or time base controls that are off their "cal" position will cause you all sorts of problems.
An oscilloscope is simply an electronic machine for drawing graphs of vaoltage against time. Most of the controls are simply to set the scale of the graph (vertical or horizontal). The controls in the trigger area enable you to easily obtain one overlaid trace, rather than a moving tangle.
In the trigger department of your scope, there should be a button marked "Line". It will be near buttons selecting "Ch1", "Ch2" or "Ext". "Line" triggers your scope from the AC mains, and is an excellent way of quickly determining whether noise is main derived (if it is, it will be stationary, if it's oscillation, it will roll through the screen).
As for calibrating your oscilloscope, there are lots of adjustments that can drift, and who's to say they've all drifted in the same direction? In other words, "If it isn't wrong, don't fix it." What you do need, though, is to calibrate the user! Either you need a good electronics engineer to show you what everything does, and why, or you need to read a book to tell you how to drive an oscilloscope. However, the first thing you can do is to look for small (front panel) pots with "cal" marked at one end, and make sure they are in the "cal" position. "Variable" or "fine" sensitivity or time base controls that are off their "cal" position will cause you all sorts of problems.
An oscilloscope is simply an electronic machine for drawing graphs of vaoltage against time. Most of the controls are simply to set the scale of the graph (vertical or horizontal). The controls in the trigger area enable you to easily obtain one overlaid trace, rather than a moving tangle.
Thanks. That's a good idea, I'll try that out.
I think I understand most of the controls on the amplifier plugin modules. On the timebase module, I'm not too sure about how the delayed triggering works or what I could use it for, but I *seem* to have the basic stuff down. For most of these photos, I was triggering off the input signal.
I have a couple of online articles bookmarked. Maybe I should go back and re-read them. I've used simpler scopes in college, but only to look at sine waves and square waves, never to troubleshoot or look for noise sources.
Saurav
Delayed time base.
Delayed time base (or "B" time base) is a faster time base that fires a controllable time after the first time base. It allows you to investigate a specific part of a waveform. It's essential for those complex waveforms like video that your college couldn't afford. On your scope, there is probably a sequence of buttons marked "A" (primary time base), "A intens" ("A" time base with intensified section where the "B" time base will be), "Alt" (alternates between "A" sweep and "B" sweep), and "B del" (delayed "B" sweep only). Try going into "B intens", and you should find you have a normal trace with one section rather brighter than the rest. Somewhere, you will find a silver multiturn knob with counter, marked "delay". Twiddle this, and you will find that it moves the bright-up (it's the delay to the "B" sweep). Just like the "A" time base, the "B" time base has its own sweep speed and trigger. The sweep speed is often adjusted by pulling out the main sweep control and rotating it. You should be able to make the bright-up wider or thinner. If you go to "Alt", you can view the expanded "B" trace while still seeing where you are looking. Once you've decided that you are looking in the right place, you go to "B del", and that removes the "A" trace to leave an uncluttered screen.
Delayed time base can be useful for investigating power supply quirks like ringing when rectifiers switch off.
Delayed time base (or "B" time base) is a faster time base that fires a controllable time after the first time base. It allows you to investigate a specific part of a waveform. It's essential for those complex waveforms like video that your college couldn't afford. On your scope, there is probably a sequence of buttons marked "A" (primary time base), "A intens" ("A" time base with intensified section where the "B" time base will be), "Alt" (alternates between "A" sweep and "B" sweep), and "B del" (delayed "B" sweep only). Try going into "B intens", and you should find you have a normal trace with one section rather brighter than the rest. Somewhere, you will find a silver multiturn knob with counter, marked "delay". Twiddle this, and you will find that it moves the bright-up (it's the delay to the "B" sweep). Just like the "A" time base, the "B" time base has its own sweep speed and trigger. The sweep speed is often adjusted by pulling out the main sweep control and rotating it. You should be able to make the bright-up wider or thinner. If you go to "Alt", you can view the expanded "B" trace while still seeing where you are looking. Once you've decided that you are looking in the right place, you go to "B del", and that removes the "A" trace to leave an uncluttered screen.
Delayed time base can be useful for investigating power supply quirks like ringing when rectifiers switch off.
You seem to be describing my scope very accurately Thanks for that information. I'd got it into the mode where one section of the waveform was brighter than the other, but I didn't know what it meant. And yes, I have a silver multiturn knob, and the main timebase control pulls out like you said.
Thanks for that information. I'd got it into the mode where one section of the waveform was brighter than the other, but I didn't know what it meant. And yes, I have a silver multiturn knob, and the main timebase control pulls out like you said.
That looks like a Tek 7603 !!
Hi,
This scope has a calibrator built in (just below the power switch)
producing 1KHz at a range of voltages. The timebase is adjusted
with an easily visible preset. Your vertical amps (7A26 ?) are
calibrated by using a screwdriver in the "cal" control by pushing in
and turning until the adjustement engages (keep pushing in).
Turn until calibrated, Dont forget to push the "cal" control back in
after. This should be plenty close enough !!
Dave /also has a 7603/7A22/7A26/7A13/7B53
Hi,
This scope has a calibrator built in (just below the power switch)
producing 1KHz at a range of voltages. The timebase is adjusted
with an easily visible preset. Your vertical amps (7A26 ?) are
calibrated by using a screwdriver in the "cal" control by pushing in
and turning until the adjustement engages (keep pushing in).
Turn until calibrated, Dont forget to push the "cal" control back in
after. This should be plenty close enough !!
Dave /also has a 7603/7A22/7A26/7A13/7B53
A picture is worth a thousand words...
That's the power of taking a photograph! I didn't recognise the model, but I could just make out a Tek logo, so I guessed how the controls would be implemented. You have a very nice scope there...
Saurav said:You seem to be describing my scope very accurately
That's the power of taking a photograph! I didn't recognise the model, but I could just make out a Tek logo, so I guessed how the controls would be implemented. You have a very nice scope there...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.