Preamp causes hum even when off
to...@fullnet.com
preamp could cause the main power amplifier to produce hum, even when
the preamp is not powered? It also causes the hum when unplugged, as
well as when powered and in use. Other signal sources do not cause
the main amplifier to hum. Other than causing the hum, the preamp
sounds good.
PhattyMo
Possibly a ground loop..
Does it hum with all of the sources connected to the preamp?
Does it go away if you disconnect the sources one by one? Perhaps it
only does it when,(say)the CD player or TV/DVD is connected.
Is the internal wiring in the preamp in good shape?
The signal carrying wires are all shielded,etc.?
Might be poor shielding,but probably not likely.
A couple cents. :-)
to...@fullnet.com
Thanks, PhattyMo.
Only one source was connected at a time. Only the preamp caused hum.
It still causes hum even when unplugged from power.
The internal ground-scheme was (and still is) pretty ugly. But now,
at least, all circuit and i/o grounds connect to the chassis through
only one point, with a 'safety disconnect' network between that point
and the chassis, consisting of 10R, 0.1uF and two anti-parallel
diodes, all in parallel. Only the earth ground connects directly to
the chassis, which is steel by the way.
Internal (and external) i/o wiring is all shielded.
Note that it is not my preamp. I'm just trying to help out. It was
built from a commercial DIY kit.
Below is a link to thumbnail links to schematics,
layout diagrams and photos, etc:
http://home.comcast.net/~garyworld/site/?/photos/
The upper-most thumbnails are the 'originals'. There are some farther
down that may show modifications, or proposed modifications, scope
traces, etc.
[Please note that the filaments' (DC) supply's smoothing capacitance
is much larger than what is indicated on the power supply schematic.]
The grounding scheme was/is quite ugly, and IMO should at least
eventually
be completely converted to a proper star-grounding scheme.
*****
BUT, my current thinking is that if we can solve the hum problem for
when the preamp is NOT powered, then we will probably have mostly
solved the problem for when it IS powered, possibly with minimal
modifications.
*****
There is an ongoing discussion thread about this preamp's problem,
where, so
far, some great minds have not found a solution, at:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=116374
(Please try to ignore my sometimes-embarrassing contributions to that
thread.)
Thanks again, PhattyMo
- Tom
Peter Wieck
> (Please try to ignore my sometimes-embarrassing contributions to that
> thread.)
Tom:
Forgiven... of course.
Tell me, is that a horked PAT-4 (Dynaco) case?
A couple of questions, although I am leaning with Phatty so far:
a) Hum is same with top on as top off?
b) Does the hum follow the volume or is it separate and always at the
same level?
c) If you short the inputs to the pre-amp, does the amp still hum?
d) Is there AC or DC on the tube filaments (heaters)?
One thing you might try: get a small screwdriver with a good plastic
handle (well-insulated). While the pre-amp is connected and humming,
but *NOT* plugged in, try touching the various shields and signal
solder points with the tip of the screwdriver. If one or more points
increases or decreases the hum, please let us know - as that will be
your point-of-departure for a search for the cause.
Please let us know.
Peter Wieck
Wyncote, PA
to...@fullnet.com
Thanks, Peter!
I hope I mentioned that it's not my preamp, and I am just a go-
between, trying to help out, while the main discussion continues on
the linked-to diyaudio.com thread..
Answers to your questions:
I am not sure of the origin of the case. Sorry.
a) The top on/off issue was recently brought up. I think that the
answer was that it was the same, either way.
b) I am not sure if the volume control affects the hum level. I think
I read that it does not.
c) With inputs shorted, hum is unchanged.
d) Tube filaments have SS-rectified and smoothed DC, except for the
12X4 rectifier tube for the B+, which has AC on filament, from the
same source as the discrete SS diode bridge's AC.
Thanks for the tip about the screwdriver. If you don't object, I
would like to post that in the diyaudio thread (including proper
credit to you, of course).
Thanks again, Peter.
- Tom Gootee
Peter Wieck
> Thanks for the tip about the screwdriver. If you don't object, I
> would like to post that in the diyaudio thread (including proper
> credit to you, of course).
How about this - you pass the tip and take credit if it works. If it
does not, blame me. It is a very old poor-man's signal tracer trick
from even before my time.
To be brutally honest with you, from what I see as to craftsmanship, I
would start over with that kit, working from the crimp-connectors
back. I would also lean towards twisted pairs ILO shielded cable,
leave one end unconnected, typically the one at the volume-control
side. I would also make sure that the board is "above ground" at its
chassis screws - use plastic screws or proper insulators if necessary.
Square off the ground connections - run them as close to the chassis
as possible to the ground point, not wildly overland as they appear
now. Try to eschew multiple grounds - jump from shield-to-shield
*then* to ground, keeping the jumpers as short as possible. As it is
now, your shield grounds appear to go from the input side to the
volume control to the board to the output side. That is an awfully
long way and ample opportunity for problems to occur. Better you
connect the shields to ground at the common ground, and run either a
twisted pair or disconnect the ground at the VC both directions.
I have to guess that all this has been suggested already?
Peter Wieck
Wyncote, PA
Eeyore
to...@fullnet.com wrote:
> Can anyone shed some light on possible reasons that a tube-based
> preamp could cause the main power amplifier to produce hum, even when
> the preamp is not powered?
Well .. when a signal source (such as a preamp) is TURNED OFF the output
becomes high impedance and it's really quite likely that hum will result
if other equipment further down the signal chain is switched ON.
I came acroso this only earlier this evening with a pro-audio sound
reinforcement setup because I'd forghotten to turn the damn mixing
console ON ! Had me going for a few minutes ! :-(
Why are you worried about hum in this scenario ? Does it also hum too
much when the preamp is ON ?
Graham
to...@fullnet.com
Hi Peter,
Thanks much.
In many of my own low-voltage solid-state proto circuits, I use a
moist fingertip, when looking for places that might need compensation,
for example. But I hadn't thought of suggesting something like the
screwdriver trick, in this case. How about if I give you the credit,
if it works, but give myself 'credit' if it doesn't? :-) It was
very generous-minded of you to suggest doing it the other way around,
though. Thank you, again. (Back when I wrote software, I used to
sometimes do that type of thing for employees of my customers, when
they'd mess something up horribly, telling them that they could tell
their boss that I said it was my fault.)
I do think that much, but not all, of what you mentioned has been
already suggested. But some of THAT might also have been overlooked
or discounted, so far, amid the initial flurries of suggestions.
But I do know that the all-metal fasteners that held the PCB to the
chassis have already been replaced with nylon standoffs.
It might be much better if he WOULD start over, and, if so, also at
least redesign the PCB's grounding scheme and fabricate a new PCB.
I must say that it is certainly a slow and somewhat difficult process,
trying to help someone fix a problem like his, via discussion-group
messages. One advantage, though (besides the more-obvious knowledge-
base-building-and-archiving types of aspects) is that the suggestions
and results can be fairly-well documented, which can be very helpful
when others join an ongoing effort, later.
By the way, I went ahead and posted a link to this rec.audio.tubes
thread in the diyaudio.com thread about this preamp problem.
to...@fullnet.com
wrote:
Hi Graham,
Thank you for posting.
Yes, it hums too much when turned on. I assumed that we should tackle
the 'hum when unpowered' problem, first. But, after reading your
post, it now appears that my assumption might have been incorrect.
Here is a quote from the preamp owner's most-recent post at
diyaudio.com:
------- quote:
measured at preamp out, preamp on:
http://home.comcast.net/~garyworld/preampon.jpg
the little numbers indicate a 2.5V 120 Hz wave, a bit modulated
(perhaps by small 60 Hz)
measured at speaker out of amp with both on:
http://home.comcast.net/~garyworld/bothon.jpg
this is a 25-30V 120 Hz wave with a more pronounced modulation by
something else.
Once I corrected the probe ground problem Tom noticed, the amp out
read flat when on and off, and the preamp measured flat when off but
plugged in.
For what it's worth, with the changes I made today the preamp hum
continues.
--------- end quote
Eeyore
to...@fullnet.com wrote:
> Eeyore wrote:?
> > t...@fullnet.com wrote::
> >
> > > Can anyone shed some light on possible reasons that a tube-based
> > > preamp could cause the main power amplifier to produce hum, even when
> > > the preamp is not powered?
> >
> > Well .. when a signal source (such as a preamp) is TURNED OFF the output
> > becomes high impedance and it's really quite likely that hum will result
> > if other equipment further down the signal chain is switched ON.
> >
> > I came acroso this only earlier this evening with a pro-audio sound
> > reinforcement setup because I'd forghotten to turn the damn mixing
> > console ON ! Had me going for a few minutes ! :-(
> >
> > Why are you worried about hum in this scenario ? Does it also hum too
> > much when the preamp is ON ?
> >
> > Graham
>
> Hi Graham,
>
> Thank you for posting.
>
> Yes, it hums too much when turned on. I assumed that we should tackle
> the 'hum when unpowered' problem, first. But, after reading your
> post, it now appears that my assumption might have been incorrect.
Performance when one or more pieces of equipment are off is an entire
irrelevance !
As to why it hums when ON is something that would involve a 'fault finding' /
signal tracing exercise.
Graham
to...@fullnet.com
wrote:
HI Graham,
I don't know that it's an ENTIRE irrelevance. But both avenues have
been explored, rather thoroughly, already. That is why I have gone
outside of the original diyaudio.com discussion, soliciting help.
Thanks, though.
Tom
Phil Allison
<to...@fullnet.com>
I don't know that it's an ENTIRE irrelevance. But both avenues have
been explored, rather thoroughly, already. That is why I have gone
outside of the original diyaudio.com discussion, soliciting help.
** You are being a complete ass.
There is NO way for YOU to solve this problem via the internet.
The owner of the gear MUST come here himself to answer questions.
He had made a REALLY DUMB mistake and cannot find it himself cos he has no
fault finding skills.
....... Phil
Patrick Turner
This is like considering how fast a chook can run after its head is cut
off.
Fast, it seems, from what you've told us.
Anyway, the preamp may not have good ground connection,
so when its off, but with its output connected to the power amp input,
you have a giant antenna effectively connected to the power amp,
and at this input source resistance is very high, and thus noisy.
When you turn the preamp on, the hum gradually fades, no?
That's because the source impedance becomes much lower,
especially if the preamp output has a cathode follower output.
I've noticed this effect sometimes myself.
So turn on the preamp first and wait until it is warm and dc voltages
have settled down which is about 40 seconds for most pres, then turn on
the power amp.
Noise with gain turned up full on the preamp line stage should be
negligible.
Patrick Turner.
to...@fullnet.com
> <t...@fullnet.com>
Thanks for responding, Phil. I love you too.
You might be right, of course. But I decided it was worth the risk.
Given the number of experienced, intelligent techs who have been
helping to try to solve this problem, and the number of suggested
tests and modifications that have already been performed by the
preamp's builder, it's getting less and less likely that the problem
was caused by a "really dumb mistake" (although we can still hope...),
or that he "has no fault finding skills". If it were that simple, I
wouldn't be here asking for help.
Neither the owner/builder of the gear nor anyone else knew that I was
going to solicit help in another group. But I have since posted this
thread's URL in the original diyaudio.com thread. So it is possible
that the owner of the gear, and/or others from there, WILL come here.
But that's up to them.
If you have no desire to be helpful, that can be your choice,
obviously. On the other hand, if you think you might be able to help,
and also want to try to help, it would be a simple matter to pop over
to diyaudio.com and contribute some of your knowledge, skills, and
abilities, rather than sitting a screen or two away waiting for them
to come to you.
Thanks again, Phil.
- Tom Gootee
to...@fullnet.com
Thanks for the analysis and suggestions, Patrick! By the way, I've
read many of your posts, and have been impressed by your experience
and accumulated knowledge.
I'm afraid this one's a bit of a hard case.
The hum does not fade after preamp power on. And much has already
been tried.
Almost the same theories as yours above are being discussed in the
original thread, right now, along with ways to try to prove or
disprove what is actually happening. See
http://www.diyaudio.com/forums/showthread.php?s=&threadid=116374 .
Wow. Now that I think about the great tube-amp stuff I've seen you
write about, I can hardly believe that you aren't on diyaudio.com.
You would be so totally in your element, there. (Hint, Hint. :-)
Thank you, Patrick. Sorry to have been so brief, but I need to get
some sleep.
- Tom Gootee
to...@fullnet.com
>
> - Show quoted text -
Some good news!!
Below is the most-recent post by the preamp's owner/builder.
----- QUOTE:
Great News...
It would seem the star grounding has worked; the 120 Hz hum is now
absent from the preamp out and amp out when preamp is off. There is
still 100mV 60Hz hum when they are on but I see this as an important
victory and feel very happy. It was previously theorized that this hum
was the sum of several issues possibly compounding each other.
Here is a pic of the current state. Once again, thanks for the
constant support and advice:
http://home.comcast.net/~garyworld/preampmarch1.jpg
Here is a close up of the starground:
http://home.comcast.net/~garyworld/starground.jpg
Please let me know if this can be improved to make it final.
Here is where you guys are going to get annoyed with me...I made the
grounding changes prescribed most recently by megajocke. I tested the
system without a successful outcome and proceeded to install the 4
(replacement) output caps and scrub all the pcb connections with
alcohol and toothbrush. Then I realized that the rca output cable was
still plugged into the old, now floating, output jack. So I can't
conclude which one of the three changes caused the improvement. I'm
assuming it's the ground scheme. Mark, I'm sorry about this. I will
gladly replace the old caps if you are interested in a conclusive
answer. Let me know.
Divide and conquer. Tom, I'm going to order resistors to step down the
scope probe signal to my computer. What is the next step?
You guys are the best. Talk to you on Monday.
gary
ps, I cleaned up the picture site: http://home.comcast.net/~garyworld/site/?/photos/
I will open up the tranny box and take pictures, I get the feeling
there are some improvements I can make with the fuse and switch
wiring.
----- END QUOTE
legg
The output structure and effective output impedance of this preamp is
lergely unchanged powered onf. It's a tube anode driving another tube
constant current source. Internal load is a 330K resistor after
capacitive decoupling, tied to ground. I'd have thought anything over
33K was overkill.
Check the schematic out.
http://home.comcast.net/~garyworld/schematic2.jpg
First-order approximation of gain's got to look like it's headed for
50dB, which it'll never reach, of course. I doubt the input volume
control ever gets past the first few degrees of travel, in practice.
Some turn-on thump, eh?
Have you any idea why anyone would pick off the output signsl from the
cathode of V301A? This has got to degrade the stiffness of the current
source. Shouldn't it be the junction of R308-9 ?
RL
legg
> http://home.comcast.net/~garyworld/schematic2.jpg
>Have you any idea why anyone would pick off the output signsl from the
>cathode of V301A? This has got to degrade the stiffness of the current
>source. Shouldn't it be the junction of R308-9 ?
In a simple model, using the Dncan Amps tube .cir files, the direct
cathode tap, as drawn, does indeed produce a dB extra gain into high
impedance loads (31.4 vs 30.7). At lower impedances, the gain
difference was more than 3dB (27 vs .23.3).
RL
Patrick Turner
Ok, I missed that it seems.
You are not alone in finding preamp A connected to power amp B
gives some hum. Hum persues me around the damn workdhop most days
and I have to "try things" to get rid of it. Often the hum is so low
I can't measure it, but its still a nuisance.
The 0V rail of the preamp should be a short buss wire about 100mm long
with ends
connected to the RCA plug 0V bodies at inputs and outputs.
ALL parts with OV connections should be made to this buss,
and the CHASSIS or CASE connected via a 5 watt 27 ohm R, and the case
taken to the
green/yellow wire to the wall socket so the case can't become live to
mains or the B+.
There should be NO direct connection of the OV buss to the case.
The hum fromj the preamp should be large enough to display on a CRO, and
you need to know if its mains F or 2 x mains F which is the rectifier F.
Here our hum is usually either 50Hz or 100Hz, and related harmonics, and
if its mainly 100Hz plus switching H,
its usually because the PSU is developing a small voltage because of
peak charge currents
in a length of the 0V buss, so you have to work out if this is tha case
by careful circuit layout.
If the 0V buss runs right close to the mains tranny, induced hum can
result.
If the preamp has been altered in its life by an audio amateur, they may
have
ruined its previously clean behaviour.
Anyway, I have had lots of itrms brought to me with hum problems, and it
takes some practised intuition to
reduce its levels below audibility.
Sometimes B+ rails are not well enough filtered or the filament heaters
are leaking ac into their
cathodes, so change to DC heaters if you can. Often changing tubes does
nothing to make it less hummy.
Centre tapped AC heater windings or with a hum minimization or hum
nulling pot
is sometimes a fix, but all so often it isn't, and ONLY with DC to all
heaters do you
fully eliminate the hum from heater problem.
> Almost the same theories as yours above are being discussed in the
> original thread, right now, along with ways to try to prove or
> disprove what is actually happening. See
> http://www.diyaudio.com/forums/showthread.php?s=&threadid=116374 .
>
> Wow. Now that I think about the great tube-amp stuff I've seen you
> write about, I can hardly believe that you aren't on diyaudio.com.
> You would be so totally in your element, there. (Hint, Hint. :-)
All the ppl in the little moderated audio groups know where I am and
read some posts sometimes. And I have a website at
http://www.turneraudio.com.au
Few in these diyer groups contribute at r.a.t for fear of getting their
heds
cut off and finishing up like the chook I mentioned above, running
around fast,
but not for very long, or very well.
Most ppl get wise to hum problems after they build amp no 2 or 3,
and they wouldn't dare ask anyone's advice; they damn well stay in their
shed until
they find the problem and discover its cause, and solve the problem.
I sure had to put in lots of sleepless nights to learn,
and armed only with theory books before I got a PC and learnt to type in
about 2001.
>
> Thank you, Patrick. Sorry to have been so brief, but I need to get
> some sleep.
Sometimes what is wrong comes to mind the day after a night's sleep.
Patrick Turner.
>
> - Tom Gootee
Patrick Turner
The 0V of the RCA appear to go to the chassis via a resistor and cap in
parallel.
That is OK, but you have two diodes also there which would clamp the 0V
voltage
to +/- 0.7V peak with respect to the chassis if ever there was any
voltage present.
So t is to be hoped that the chassis voltage and 0V rail
do not have any hum voltage above say about +/- 0.4V pk lest the diodes
begin conduction and allowing noise into the 0V rail.
>
> Please let me know if this can be improved to make it final.
>
> Here is where you guys are going to get annoyed with me...I made the
> grounding changes prescribed most recently by megajocke. I tested the
> system without a successful outcome and proceeded to install the 4
> (replacement) output caps and scrub all the pcb connections with
> alcohol and toothbrush. Then I realized that the rca output cable was
> still plugged into the old, now floating, output jack. So I can't
> conclude which one of the three changes caused the improvement. I'm
> assuming it's the ground scheme. Mark, I'm sorry about this. I will
> gladly replace the old caps if you are interested in a conclusive
> answer. Let me know.
>
> Divide and conquer. Tom, I'm going to order resistors to step down the
> scope probe signal to my computer. What is the next step?
>
> You guys are the best. Talk to you on Monday.
>
> gary
>
> ps, I cleaned up the picture site: http://home.comcast.net/~garyworld/site/?/photos/
The schematics look OK. but use careful examination to make sure
EXACTLY what earth paths exist; there could be
a direct connection between 0V and the chassis that you are not aware
of.
Patrick Turner.
Patrick Turner
My first tube preamp had a very similar circuit to yours. Its SRPP.
Mine used Rk for top and bottom triodes of 1.5k, not 680 like you have.
The top triode presents a load to the bottom triode which is NOT a CCS,
but an active resistance = Rk ( µ + 1 ) = approx 680 x 18, or
not a very high R at all. Rout from the top cathode is probably only 2k
ohms.
But this circuit works well enough with 1vrms output, more than enough
to
drive most power amps.
The heaters for the top and bottom tubes should be dc, and this dc
supply should be
biased at +75Vdc and bypassed to 0V via 100uF.
Its where hum will definately get in if you have ac heaters!
>
> First-order approximation of gain's got to look like it's headed for
> 50dB, which it'll never reach, of course. I doubt the input volume
> control ever gets past the first few degrees of travel, in practice.
>
> Some turn-on thump, eh?
>
> Have you any idea why anyone would pick off the output signsl from the
> cathode of V301A? This has got to degrade the stiffness of the current
> source. Shouldn't it be the junction of R308-9 ?
Current sources ain't stiff.
The output resistance at the top of R309 is higher that at the top of
R308, the top tube cathode.
So the outside world load disturbs the amp more if connected to the top
R309.
The schematic has 220 ohm oscillation stopper resistors which is fine,
but I've never found them to be needed,
although the output 220R, R311, could be 470R.
See my website pages on preamps for more details about line stage
preamps and the use of TRUE CCS dc supplies.
http://www.turneraudio.com.au
Patrick Turner
>
> RL
memja...@yahoo.com
> 50dB, which it'll never reach, of course. I doubt the input volume
> control ever gets past the first few degrees of travel, in practice.
>
> Some turn-on thump, eh?
>
> Have you any idea why anyone would pick off the output signsl from the
> cathode of V301A? This has got to degrade the stiffness of the current
> source. Shouldn't it be the junction of R308-9 ?
>
> RL
Hi, Gary here, preamp's owner. RL you are absolutely right, I have
never needed to turn it past 9 o'clock and there is a lot of dangerous
transients on warm up/turn on.
legg
>
>Have you any idea why anyone would pick off the output signsl from the
>cathode of V301A? This has got to degrade the stiffness of the current
>source. Shouldn't it be the junction of R308-9 ?
>
Mr Babiak sent a link to this article covering the tap choice/effects
and expected gain in this type of circuit.
http://www.tubecad.com/may2000/
RL