Protek P-3502C 'Scope - Broken?

[sluckey]

So to measure B+ ripple on the B+ rail, just set the meter to ACV, one lead on the B+, and the other on ground?

Depends on the meter. It works well with a Fluke 87. My analog Triplett cannot measure ripple in this manner. So, to be sure, check your owners manual.

Look at this schematic... http://home.comcast.net/~seluckey/rocky/rocky.pdf
I just now measured the ripple and the values are 8.6VAC for node A, 120mVAC for node B, and 6mVAC for node C. These numbers may not be very meaningful on their own, other than to show how effective the filters are. Notice how fast the ripple decreases as you add more RC networks.

These numbers really become useful if you will write the ripple voltage on the schematic when the filter caps are new. Also make note of what model meter you used. Then you have a baseline to compare when you are servicing the amp years down the road. Just recheck the ripple and compare to the baseline readings. You'll know if the caps are still doing a good job.

Also, if you work on many different amps and regularly measure ripple voltages, you will soon develop a 'feel' for what's acceptable even for an unfamiliar amp.

[PRR]

> I presume meters have a blocking cap built-in somewhere...

As Sluckey says: not all. In particular "passive" VOMs often use an "AC" scheme that reads both AC and DC. Some provide an externally jumperable cap to block DC, many don't.

But even if so: what rating?

My Heath VTVM had a 400V(?) cap. I managed to blow it. Of course a HeathKit is totally repairable. I fixed it with a 1,600V cap. Haven't blown that yet.

But these new meters may call 100V "high voltage". And if you blow it, you may not be able to find it.

Read The Specs.

> don't probe any place in the amp that has more than 300VDC present.

Technically, 300V DC plus 10V peak ripple is 310V, and could be 310V against the 300V rating, which is not allowed.

True, ripple in preamp stages generally must be well under one Volt. And there ought to be some safety margin on the rating; 301V should not be fatal.

Still in all, it is better to use your brain and lever-down. A 1Meg and a 10K resistor from B+ to ground will give 100:1 leverage. The B+ will read like 3V, 10V of ripple will read as 0.1V, and nothing is very stressed. There's little errors this way but around 1%, close enough for jazz. (I know why Sluckey went direct: we were debating a 1% effect, so he wanted to avoid additional differences even at the risk of popping a 'scope input. I think the moral is: don't sweat the small stuff.)

[jjasilli]

OK, so is this right:  100:1 voltage divider -- tap-out between: 1M series resistor & 10K shunt resistor. 

Voltage drop across 1M series resistor  = 396V; Watts = 396²/1M = .16.  1/2W resistor is OK?
Voltage drop across 10K shunt resistor = 3.96V; 3.96²/10K = .0016.        1/2W resistor is OK?

[jjasilli]

With the help of some vodka and a very nice man with an oriental accent on a youtube video, I now think that maybe I realize that:  my scope is very slightly out of calibration; the value of the graticule divisions changes with the volts/div knob setting, and probably the time/div knob; DC is easier to measure than AC (but I still get both wrong); "trigger" is the name for freezing the waveform (and not Roy Roger's horse); you need a probe adjusted and labeled for ea scope channel because they'll be a tad out of whack; if I'm ever to finish my amp project I need to lay off the 'scope, or just use it to check general waveform shape.

The Tone Lounge mentions some reference books.

[sluckey]

I think you're ready to step up to Tektronix... And Jack Daniels!