High VAC on B+

[CrocMule]

I picked up a Danelectro DS 50 cheap in "non-functioning" condition from a woman who was using it as a side table.  I carefully started it up, and it actually works.  The reverb sounds like trash, and there's an annoying hiss I can't track down yet -- perfect static hiss, like a blank D-90 turned up loud.  But the amp has good tone, if you're looking for twang -- surprising bass too.  Strat quack sounds wonderful.

All the ecaps are new F&Ts.  One of the F&T 100uF caps came with a small dent, but it tested good, so I used it.  I also replaced the bias and rectifier diodes with 1N5408s (what I had) and the 68 Ohm 6L6 screen resistors.

There's a slight 120Hz hum/buzz, but it's faint in the hiss.  I'll save the hiss issue for later, since there's something potentially bigger going on.

I was checking voltages, and noticed high voltage AC on the B+ lines.  My DMMs wouldn't lock on to the voltage, so I changed the batteries in my old analog VOM and tried that.  According to that, B+4 has ~1040VAC on it.  I replicated that measurement several times and compared it to wall voltage on the same 1000VAC range.  I don't know if it has any oomph behind it though -- could a phantom AC voltage be more than twice a DC voltage?

Here's a chart of readings with DMM and analog meter.

               DMM VDC          DMM VAC          VOM VDC          VOM VAC
B+1         114                   0.3 - 34            ~110                ~225
B+2         229                   2.2 - 255          ~224                ~475
B+3         344                   3.4 - 281          ~319                ~735
B+4         460                   2.6 - 309          ~430                ~1040

The VAC readings on the DMMs jump around in a repeating cycle.  Each B+ node hits OL in the cycle (the DMMs are rated at 600V).

I'm not sure what to do at this point.  The diodes all test good, and the ecaps all ESR good.  Any ideas?  Suggestions?

A hearty THANKS in advance to any help you guys can provide!

[CrocMule]

Here's a gut shot

[sluckey]

Probably nothing to worry about.

Many cheap DMM meters cannot  properly read AC volts riding on a DC voltage. I've never seen an analog VOM that could properly read AC volts riding on a DC voltage.

So, what are the brands, model numbers, and approximate costs of your meters?

[CrocMule]

Probably nothing to worry about.

Many cheap DMM meters cannot  properly read AC volts riding on a DC voltage. I've never seen an analog VOM that could properly read AC volts riding on a DC voltage.

So, what are the brands, model numbers, and approximate costs of your meters?

That's good to hear.

They're cheap.  You'll get a chuckle out of the analog -- Sears from about 1980.  It was my first piece of test gear -- asked for it for Xmas.  It's good for tremolo voltages.  The old Extech 430 was about $50 20 years ago.  The new one was $80 a couple years ago.  The B+K 316 was maybe $160 10 years ago.

Thanks
Russ

[PRR]

That "Sears" was a good meter for the time and price. I have the same machine with a different brand sticker.

I don't know why you should try to measure a few V AC in the presence of 500V of DC. That's like seeing Venus at noon; the glare makes it hard. If you have major AC on B+, your ear will know.

And since this was an incredibly cheap amplifier new, it may have buzzed fresh off the truck and you may have to "over-restore" it to get buzz down to modern tastes.

[sluckey]

That is a decent VOM. I still prefer to use one when checking diodes and transistors. Like seeing that front to back ratio. Checking caps was neat on the ohms scale too, especially electrolytics. Nothing like connecting a cap and watching the needle quickly jump toward zero, then drifting back to some high resistance. If the needle stopped then started creeping back toward zero, you got a bad cap.

Are you an electrician or HVAC tech? Not many people have a need for a clamp on meter.

Here's my first meter. I bought it from Radio Shack in '68. Served me until I got a Simpson 260 in '74. Still have both and still in working order.

[CrocMule]

If you have major AC on B+, your ear will know.

Thanks. I was kinda suspecting as much.  I was more surprised and cautious than really alarmed by the 1040VAC reading. It didn't make sense. 

The buzz isn't bad, at least with the hiss going.  It's a constant steady-state hiss, smooth white noise, unaffected by any of the volume or tone controls.  Pulling V1 (reverb) and V2 (input) does nothing to the hiss (or V5 (tremolo)) -- the hiss is still strong with only V3 and V4. But pulling V3 or V4 silences the amp.  Swapping V3 or V4 for known good tubes does nothing to the hiss either, it's still strong and unchanged.

As I said, all ecaps are new F&Ts (except for the one 25@25 bypass cap which is a Sprague).  They all tested over 20M resistance, capacitances were all within tolerance, and they ESR'd good.  After installation they all ESR good.  I'm not sure where to go next, other than to test all the carbon comps.  I guess that will be the next step.

Attached are voltages.  They all look fine except on V1 -- they're way low.  Swapping the two 6FQ7s doesn't change the voltages much.

A NOTE ABOUT MY VOLTAGE READINGS: I live off grid with a sine wave inverter that is stellar in every way but voltage regulation.  The wall voltage wanders around constantly which makes accurate voltage measurements impossible.  When I take a reading I glance at the Kill-a-Watt meter I keep plugged in and wait until it dances around 115.5VAC, then I look at the DMM.  Usually wall voltage is between 115.0 and 116.5, but sometimes it goes up to 120 and hangs there a few minutes before wandering back down to ~115.5.  I've called the company, and that is within spec for that model -- nothing can be done to change it. 

My plan is to install a 2nd inverter -- an Exeltech with very clean pure sine and good voltage regulation for sensitive loads.  But until then, I'm stuck with the wandering MagnaSine.

You're certainly right about quality -- the amp was made with super cheap materials and components -- no penny was wasted.  It's a good thing that it's in good condition, because if it wasn't, it wouldn't be in any condition at all -- it would have disintegrated.  The particle board isn't even 1/2". 

But it's a curious amp -- lot's of character.  Three 8 Ohm (10" Jensen) speakers in parallel?  What's that about?

[CrocMule]

Are you an electrician or HVAC tech? Not many people have a need for a clamp on meter.

Electrical contractor -- have a small solar PV business, mostly off-grid.

My first electrical "project" was to disassemble and reassemble my friend's mom's hairdryer when she was in the kitchen when I was about 10.  When I turned it on sparks flew like the 4th of July -- I was hooked!

Good to know the old Sears is a decent meter -- that's funny. 

[CrocMule]

I don't know why you should try to measure a few V AC in the presence of 500V of DC.

Well, one thing led to another.  The schematic shows 185V for V5 P1 (12AX7), but I couldn't get a reading with my DMM.  So I flipped it over to VAC out of curiosity and couldn't get a steady reading on that either.  So I got out the VOM and noticed the wavering DC voltage and realized it was the trem tube I was looking at.  Duh.  But the extreme VAC reading puzzled me, hence the question.  It was ~1040 VAC, over the top of the 1000V range.  Seemed impossible.

Thanks for clarifying

[sluckey]

Probably need to step up to Fluke if you want to measure ACV_rms on a tremolo oscillator plate. 

[Lectroid]

Good to know the old Sears is a decent meter -- that's funny.

I have several meters now including my Klein daily driver.  But there on the shelf with them is my first, an old Craftsman digital from around 1985.  Strong as ever.

[CrocMule]

Probably need to step up to Fluke if you want to measure ACV_rms on a tremolo oscillator plate. 

Yup.  People have been telling me to step up to Fluke for 20 years, but I always have better things to spend money on (like a Danelectro end table).  I've never needed that level of precision in my professional work.  Even with a Fluke, I'd use a cheapo for everyday bang-about work.  You're right.  I'll look at my finances and see if I need to spend any more money this year.

Thanks

[CrocMule]

Checking caps was neat on the ohms scale too, especially electrolytics. Nothing like connecting a cap and watching the needle quickly jump toward zero, then drifting back to some high resistance. If the needle stopped then started creeping back toward zero, you got a bad cap.

Interesting -- that does sound more satisfying.  I'll try it next time.  Thanks

[sluckey]

Checking caps was neat on the ohms scale too, especially electrolytics.

Just set the meter to the highest ohm scale and connect the leads to a capacitor. You're not looking for any specific ohm reading. You're only interested in the needle movement. When you first connect to a cap the needle will quickly swing toward zero. Heck, it may even peg the needle past zero for a short time. The needle is showing the charging current of the cap. As the cap continues to charge, current decreases, and the needle will start slowly returning to some high resistance reading. The needle may even stop. This is all OK. But if the cap reaches some max resistance reading and then begins to creep back toward zero, this indicates a leaky cap. Replace it. And of course if the cap reads some low resistance or even zero, it's bad.

This is one of those tests that you need to have a good feel for how a good cap behaves. So test a bunch of good different values until you have a feel for a good cap. Doesn't take long.

And remember, the actual readings don't mean that much. And know that different meters behave slightly different. But that ole analog needle is good for a lot of stuff other than a static reading. Probably one more generation and this tip will be lost forever.

[CrocMule]

Checking caps was neat on the ohms scale too, especially electrolytics.

Just set the meter to the highest ohm scale and connect the leads to a capacitor. You're not looking for any specific ohm reading. You're only interested in the needle movement. When you first connect to a cap the needle will quickly swing toward zero. Heck, it may even peg the needle past zero for a short time. The needle is showing the charging current of the cap. As the cap continues to charge, current decreases, and the needle will start slowly returning to some high resistance reading. The needle may even stop. This is all OK. But if the cap reaches some max resistance reading and then begins to creep back toward zero, this indicates a leaky cap. Replace it. And of course if the cap reads some low resistance or even zero, it's bad.

This is one of those tests that you need to have a good feel for how a good cap behaves. So test a bunch of good different values until you have a feel for a good cap. Doesn't take long.

And remember, the actual readings don't mean that much. And know that different meters behave slightly different. But that ole analog needle is good for a lot of stuff other than a static reading. Probably one more generation and this tip will be lost forever.

That's a good trick -- I'll pass it on.  Can it be done on caps in-circuit?  Probably depends -- not with a bypass cap, or a filter cap with a bleeder.

I don't know what good ecaps I have in stock.  The two 2019 Sprague Atoms I pulled out the other day failed ESR miserably.  I was too disappointed in my sudden $40 loss to want to check the handful of others.  Those first two are in one of the junk boxes by the trash can, so at least I can get a feel for how bad caps behave.

Thanks!

[PRR]

...the extreme VAC reading puzzled me, hence the question.  It was ~1040 VAC, over the top of the 1000V range. 

I have seen that. Many meters have a loose concept of AC or DC. Many are half-wave AC. So it "might" be adding-up AC and DC values.

I had a Heath VTVM "hardened" to read 1500VAC @ 10Hz. Grossly oversize blocking cap and some rectifier tweak. All on a 1/4 plug: even with Heath modifications this was pushing our luck.

[HotBluePlates]

... B+4 has ~1040VAC on it.  ...

I assume your meters are good.

When I encounter a reading that doesn't make sense, I assume I made a mistake somewhere & try to find what that mistake is.

I see two voltage-doubler supplies stacked on top of each other to make around 500vdc (which perhaps sags to 470vdc in the schematic).

If both doublers have 1 cap charged to its DC Volts, then the next half-cycle charges the other cap to make the stack reach ~500vdc.

Some half-cycle following after that will be "-500v" if you're connected to the PT wiring somehow by accident.  That is, you're looking at the "about negative-500v AC peak" blocked by the rectifier diodes, and then seeing it added to the "about 500v DC" across the stack-o-caps.

So I suspect you had your meter leads connected in a manner that saw the "DC output volts after the rectifiers" plus the "negative AC volts before the rectifiers."