ampage
Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!

 
Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Backround Noise


 
1/26/1998 10:12 AM
Joel
email
Backround Noise
Ok, maybe another stupid question.  
 
Can a tube amp be made dead quiet at all volumes (i.e., with no guitar playing into it)?  
 
For instance, if I turn up my home stereo (transistor) with no input, there really is no noise. Can or should this happen in a tube amp?  
 
I have a Traynor Mark III, circa 1972, and I get a backround hiss/noise at all volume levels, even with the preamp and master volume off. This noise is quiet, but noticable. The noise increases when the either or both volumes are increased. The noise is not noticable when the guitar is played thru. As I said it's just in the backround. I recently changed the filter caps and noticed maybe a tiny decrease in the noise but a definate tightening up of the amps response.  
 
I changed the power tubes a few months ago and that did not change the noise. I've checked the bias and it's correct. Preamp tubes have also been changed and swapped.  
 
The Traynor has no choke. Would adding one in help?  
 
Should I change all the caps in the amp?  
 
Should I leave well enough alone? The amp works fine other than the backround hiss.  
 
Thanks in advance.
 
1/27/1998 1:16 AM
David S.
email

Joel,  
 
I'm not the expert on noise but I will share what I know and what I think I know and ofcourse, my opinion (worth about $.01 .....).  
 
I hear this same noise in all my tube and transister amps alike but it seems more pronounced in my tube amps. I have added pots in the feed back loop between the 820 ohm resister and the phase inverter to provide a damping control. When I set the pot for max feed back the hiss level drops. I think this is due to the lower gain that more feed back produces. This leads me to believe that most of the hiss is coming from the preamp tubes themselves. The "shot" noise (the electrons hitting the plates) along with Jhonson noise (noise coming from current flowing through resisters) and the high gain of guitar amps combine to make alot of hiss or (rush as I call it). I don't think that more fitering in the power supply will help this. And I would'nt worry about it unless it just makes life not worth living.  
 
I guess that If a person wanted to they could reduce the noise to an absolute minimum but, I think this would involve some special technics.  
 
Well I think I got something right... I think!  
Yea, I know I got my opinion right......  
 
David S.
 
1/28/1998 12:27 AM
Joel
email
Re: Backround Noise/DC Filiment Supply
Thanks David,  
 
I went into the amp and redid the filter caps 'a la' Dan Torres (from his book) with resistors in parallel as well as .1mfd poly caps in parallel. I also changed a preamp filter cap that I hadn't noticed before. The amp tightened up and quieted down, although that backround hiss is still present (but maybe a bit quieter). I think it must be in the power amp stage because it's audible with the volumes turned to zero.  
 
I wanted to try something else in Torres' book: he describes amp noise may also eminate from unbiased AC supply to the filiments/heaters. He recommends changing this to a DC feed. I tried to make the DC feed (bridged rectifier/big ripple filter cap/ground via .1mfd poly cap) but my measure for DC output voltage was 8.5v - too high. I checked the source filiment AC from the output transformer and that measured 6.8v. In Torres' book he described the voltage step-up thru a bridged rectifier, but said that this was theoretical and the real output should not exceed about 6.8vDC from a 6.3vAC source. The rectifier that I used was a 25amp, 600vPIV NTE thing. Was it too efficient?  
 
Needless to say I didn't hook up the high (8.5vDC) feed to the filiments for fear of burning my tubes out.  
 
Any thoughts?  
 
I think I'll try the filiment bias mod (potentiometer off the center tap of the filiment) next.  
 
Thanks in advance.
 
1/28/1998 12:32 AM
Joel
email

Actually, I just thought of this. The ripple filter cap from DC + to - was recommended to be between 4700 - 15000 mfd, as I said before, big. I used a 10000mfd 450v electrolytic cap. Could this be the reason for the step up to 8.5 vDC?
 
1/28/1998 1:40 AM
David S.
email

Joel,  
 
As you said, when volumes are turned down you still have the hiss but even with the volume down your power tubes are still amplifying the second half of each preamp triode, reverb recoverery stage and phase inverter. There gains remain unaltered. I'm not saying that the power amp stage does not contribute to the hiss level. It's just very small compared to the preamp stages which has been multiplied many times by the time it gets to the power amp. Due to all the stages that are still contributing to the hiss (even with volumes down) you can't get a accurate perception of power amp noise unless you rid the system of all previous hiss by pulling all preamp tubes out of the amp or shorting the signal path to ground.  
 
As far as the heater supply goes... When AC is rectified it will charge the caps to the AC's peak voltage which is 1.414 times higher than the AC-RMS value. This will give you about +8.9Vdc.  
The rectifiers are going to drop about a volt or two. I think the reason for your voltage being to high is that you did'nt have a load on the circiut. When the heaters begin taking current from the system the voltage will drop due to the caps being emptied when the rectifiers are turned off every cycle. This is where larger caps come in to play. By increasing the caps size it will have a larger energy reserve to dump during rectifier turn off. Go ahead and add the tube heaters to the supply and watch the voltage drop as you plug in more heaters. By rectifying the AC voltage you also will need up to 40% more power from the transformer due to losses in the rectification. I have seen some rectifyed supplies, once load with all the tubes, not be able to produce the rated 6.3 Volts due to these losses. This mod is going to lower the hum level and it may drop the hiss level a very small amount. If hiss reduction is of prime importance to you. I would start with a batch of preamp tubes and test each one for lowest noise. I think you will have better results with this method. I don't want to discourage you from your pursuits. I just want you to know the options. Hope I've been some help.  
 
David S.
 
1/28/1998 7:52 AM
art
An easier alternative to trying to use DC for  
all the filaments (which totals more than 6 amps  
for the 4 EL34's plus the 6 other tubes), is to  
just supply DC for the preamp tubes, since that's  
the critical part of the circuit anyway.  
 
I too have a traynor mark III; ten years ago  
everybody thought they were crap, now they're  
valuable antiques. I was always mystified why  
everybody seemed to change their tubes so often  
while mine lasted forever-- it's a much more  
conservatively designed circuit than, say, a  
Marshall...  
 
I agree with david, if you have noise with the master volume on '0', that really narrows it  
down; test everything in the phase splitter  
and try swapping the tube. I had a problem once  
(different amp) that turned out to be a 470k  
resistor that had drifted to over a megohm.  
 
-art
 
1/28/1998 1:25 AM
Bruce
email

Hi Joel, Bruce here...  
First off.....  
Anything your ears are telling you is a hiss sound is going to be at frequency above 2000Hz and up to or over 12,000Hz.  
Whether or not your guitar speaker can reproduce that 12KHz frequncy with any real sound pressure  
is also debateable.  
So, in my opinion, messing around with the power supply to tame the noise you hear as a hiss, is a waste of time.  
That frequency of noise is in the preamp stage and is actually additional signal noise, not power supply noise.  
Now as for the DC filaments.  
If you are measuring an RMS voltage of 6.8vac under a load then you need to drop that down a little bit.  
The diodes in that rectifier block you installed will drop the ac down by 1.2 votls when rectifiying.  
So, with a capacitor filtter, you really have  
6.8vac - 1,2v= 5.6dc x 1.4142 = 7.92 vdc,  
The caps will charge close to the peak AC voltage... not the RMS voltage.  
Still too high for the filaments.  
From here on I am going to round off numbers.....  
This is for an educational purpose and since you are doing the voltage measuring, you'll have to plug in the real numbers.  
You need to loose a little less then 2 volts here.  
How many tubes at what current is being used by this DC filament power supply?  
Are just the preamp tubes drawing from the DC filament supply or are all the 6 volt tubes getting fed the DC?  
If you are using, say, three 12AX7 type tubes then the filament suppy is drawing about 1amp ...(900ma) from the rectifier/filter.  
Ohms law is pretty easy from here.  
Loose 2 volts drawing 1 amp = 2 Ohms.  
Aproximatly ... 2V/1A=2 Ohms.  
Actual is 1.78ohms... good luck finding that value.  
That is how much additional series resistance you would need to get the filament voltage down to around 6.0 volts.  
I would use two 1 Ohm/ 2 watt resistors though.  
One in the plus side and one in the return to the rectifier/filter cap.  
That would still be 2 ohms total aditional series resistance in the DC filament supply.  
Oh yeah.... the power of the resistors.  
1 amp x 2 volts dropped = 2 watts.  
Use a 1ohm/2 watt resistor in each leg and then re-measure your DC voltages on the tube socket.  
Itl might read a little low under a load. So you will have to fine tune the components.  
Don't forget, a big 6 amp 50 v diode will also drop .6 volts across it's junction if you need to loose half volt.  
Or two diodes to loose 1.2 volts. Get it?  
Or one 1 ohm/ 2 watt resistor with a big diode in series.  
There are other way to drop this a bit.  
 
Bruce
 
   Page 1 of 3 Next> Last Page>>