Is this a good way to check ripple with a meter?

[frank57]

This is related to this other thread:
http://www.el34world.com/Forum/index.php?PHPSESSID=8ee6761176f51a5806510b9342f70e6b&topic=11014.0
But what I did was the following:
Did I do this right for the ripple?
I soldered one leg of a 102pf regular cap to the + of the second filter cap.
The other leg to a wire with a connector so I could touch the red probe of the meter to it away from everything.
On ac I got 0.175 to 0.180.The black probe I had on the ground bolt.
The dc voltages are below.

[sluckey]

Just set your meter to read AC volts and stick the probes across a filter cap.

[jjasilli]

Just set your meter to read AC volts and stick the probes across a filter cap.

I suspect that ripple may be less on a tube's plate than at it's filter cap node, because the B+ dropping resistor(s) (and plate resistor??) are part of the filtering circuit.  If you want to block hi-voltage DC to measure ripple, my vintage manuals say to use a 1uF (presumably non-electrolytic) cap in series with the trest probe.  I believe FYL has recommended .22uF for this purpose*.  But your pF blocking cap is so small that it may be at least partially filtering-out 120Hz ripple, thus giving faulty measurements.

*  Perhaps .22uF is fine for modern DMM's. My vintage manuals are for VTVM's which have parasitic capacitance issues, and require a shielded test lead, blah, blah.

[frank57]

I'll see if I have any caps that size.
I'll also try to do like Sluckey says and just use the probe.
Maybe I'll push the caps up a bit so I can get at them from the top of the pcb.

[frank57]

Well I put a 105pf (1 uf)cap on a perf board and ran a wire to the pos of the filter cap
and then touched the red probe to the other leg on the perf board.
I get a jumpy 0.185 to 0.2.
I take it this result is still not right unless it gets multiplied?
Maybe I'm doing something wrong with the meter?
I'll try to push up some of the last filter caps so I can measure directly on the leads.

[jjasilli]

105pf ≠ 1uF.  100pF = .0001uF.  http://www.el34world.com/charts/Charts.htm

[LooseChange]

When testing, what AC voltage would be considered within normal limits?? I'll assume the best would be no AC but I can't see that happening.

[jjasilli]

According to my vintage Sylvania VTVM manual (available free online), power supply ripple should be under 1%, determined as follows:  1.  measure DC volts across the output terminal of the filter; record this as E₁; 2.  do the same with the meter re-set to read AC volts; record this as E₂; 3. Ripple percentage = E₁/E₂ X 100.

[sluckey]

105pf ≠ 1uF.  100pF = .0001uF. 

That's true. But, I have never ever seen a 105pf cap. Now I have seen plenty of caps marked 105. Aint got no 'pf' following the 105 though.

The value of a cap marked 105 is 1µF. The value is determined like this...

1 is the first digit.
0 is the second digit.
5 is the multiplier, ie, add this number of zeros to the first two digits.

So, 105 becomes 1000000, or 1 million. The base unit is always pica Farads for this code scheme. So, 105 is 1µF, and 104 would be 0.1µF, or 473 (just saw this value on Plexi50's warbler build) is 0.047µF.

The code scheme is similar to the STD resistor color code.

[frank57]

I goofed it just says 105 .
When I measure it I get something like 993uf.
So first I measure the dc on the lead and then the ac?

[jjasilli]

I goofed it just says 105.  Yes the nomenclature is tricky:  105pF means 105pF (which would be an unusual value as you say).  Plain old 105 means 1uF as you say.

When I measure it I get something like 993uf.  Wrong decimal place?  Should be 0.993uF?

So first I measure the dc on the lead and then the ac?  As sluckey says above there is no need to use the additional DC blocking cap in series with your test lead, when a VTVM or modern DMM is your testing device.  (This is also stated in the Sylvania Manual I quoted above.) 

But older VOM's can be insatntly damaged by an overvoltage condition.  Also, a blocking cap may be a desirable safety step for an Oscilloscope when it is the meauring device, because the DC B+ voltage may exceed the spec of the scope's input.  Of course, you could use an X10 scope probe, but if you goof you coud burn out the 'scope.

[frank57]

It's .993 alright.
Should I leave the blocking 105 cap there or try to measure directly on the caps?
Will it affect the reading if I leave it?
It's a bit tricky to measure directly.
I think I can pull it off on the last caps.
I'll try to push the caps up a little.
But the hum and buzz definitely stop when I pull v2 out.

[sluckey]

What do you hope to gain by having a voltage number to associate with AC ripple? The number will probably be meaningless to you unless Marshall printed the ac ripple voltages on the schematic for comparison (not likely). Ripple voltage can be 10 to 50 volts on the first filter cap depending on the load current. And by the time you get to the end of the B+ rail, ripple will probably be much less than 1 volt.

[frank57]

I'm sort of hoping something unusual will pop up somewhere I guess.
Worth a try?

[jjasilli]

But the hum and buzz definitely stop when I pull v2 out.  Hum and buzz is probably not ripple. 

[frank57]

I don't know if this is part of the problem but,
looking at the ground pads for the 68uf caps, it's clear that part of the pad has been gouged out.
85 percent is still there, but I don't think these pads were designed for snap in caps or they were poorly installed to begin with.I'll do a quick check of the ac voltage and ripple stuff and then maybe try a 100uf cap.
But a big part of the problem maybe is the 100k mid pot. With the resistor it's 50k, but on a typical Hiwatt I think it's more like 18k with the 22k resistor in front of it.If you cut it down by changing the 100k resistor you definitely bring down some of that hum.
I wonder too if 470k for r8 is a smart choice.
On the Hiwatt Overdrive lead they use 47k or a 500k pot there.
Maybe 470k is way too much?
There were a few oddball hiwatts with a regular tonestack too.
I'm going to try to see if I can find a jet city 20 watter that doesn't have a hiss problem too.

[jjasilli]

More nomenclature issues, but it's the exact type of Noise which is important.  Hum & Hiss are both Noise, but they sound different, and are associated with different possible electronic problems.  Hum is low pitched and will probably be either 60Hz or 120Hz.  It's imporatnt to pin-down the octave of the hum.  60Hz is asociated with the heater supply; or maybe half-wave rectified B+.  120Hz hum is associated with the B+ supply (full-wave rectified).  (It's a B-flat on your A string, regular guitar)

Hiss is like white noise; higher pitched than hum.  Do you have hiss, hum or both?

[frank57]

I don't hear a hiss problem.
The jet city amp I had definitely had a hiss problem as you turned up the master.
There is buzzz and hum on this Hiwatt amp that goes up as we turn up the gain.
Hum goes up with the master and the mid.
The mid is contributing a lot of hum.Just using the efx loop seems to be ok.
A little bit of bzz hum.
It's there with nothing plugged in as well.
At low levels it's ok. Gain low master up to maybe 12.
Master low bzz is ok.
But if you turn up the master a little more with the gain high: bzzz.
The hum I believe is 120hz.
It seems to match b flat on the a string.
Shorting some of the parts to ground in between v2b and v2a kills it down.
I'll make a sound file and save it as a quicktime movie so you can hear it.
It's better than it was so I'll make a new sound file rather than post an old one somewhere.

[frank57]

I tried 100uf as the second filter cap and it seems to be reducing the buzz  somewhat with nothing plugged in.
The hum I don't think is affected as much. Amp overall sounds about the same slightly more bass.
The voltages are 379-380 with a bias about -16.
Definitely the mid pot is putting the hum over the top.
It's too strong at 50k I think.
When I plug in though same old bzzz if you turn the master up with the gain high.
It's not the guitar, which is fine on other amps.
Bit tough to figure out if it's 60 hz or 120hz.

Any way to figure out if the ground pads on the  68uf caps might be part of the hum problem?
On one of the 22uf caps I think the ac is .053 and the dc is 296.
I'll try to get a sound file up.