| bobhayes |
Hi all,
I've been listening to my 2A3 amp for about a year, and love it. But I recently plugged my TV into it to watch a film, and the hum is unbearable for movies because of all the silence in films.
I just wondered what the typical hum figures are for these types of amps.
I have 10.8mV left and 10.6mV right at the speaker outputs, when loaded. Is this more than should be expected from sovtek 2A3's?
I know its originating in either the 2A3's or the output transformers and reckon this is typical filament hum, so I thought I'd just check.
It's 100Hz and is there whenever I switch the power on. It remains when I pull the rectifier, so it must be the AC filaments right?
It also remains when I pull the driver tube, and only disappears when I pull the 2A3's. It is an octave above the 50Hz hum I get when I wind the hum pots to one end.
Any information would be greatly appreciated, if only to put my mind at rest.
Thanks, Bob |
|
|
| Bas Horneman |
Hi Bob,
I have read that 2mV hum on your output is attainable even with AC heated filaments.
Does your filament supply have a hum pot?
Regards,
Bas |
|
|
| bobhayes |
Both channels heve hum pots.
I can minimise the 50Hz hum, but am left with 100Hz hum remaining both channels.
I read through some other posts but did not find the exact problem I described above. |
|
|
| SvErD |
Do you have cable-TV? Does the hum go away if you pull out the antenna-cable?
Then you need a galvanic isolator.(Balun?) |
|
|
| Bas Horneman |
| quote: | | 100Hz hum remaining both channels |
Probably a ground loop somewhere...since PSU ripple hardly ever is a problem. Is it a homemade amp? |
|
|
| bobhayes |
Thanks SvErD,
The problem is the same whatever the source, or even without one, so I don't think the TV is the problem.
Bas Horneman,
Yes, the amp is home built. Where should I look first for ground loops?
It's such a simple circuit (JE Labs/Angela 6SL7/2A3) and as far as I can tell there should be no ground loops.
It is however, built on an aluminium chassis. Could the ground plane resistance (one ohm end to end) play a role in this?
Thanks |
|
|
| Bas Horneman |
Hi Bob,
I'm still learning about grounding..it's not easy. What does your current grounding scheme look like? Star ground has worked well for me...except in the last amp which I built for someone else...someone else took over and fixed my mess... and made a seperate star earth for signal and psu and joined the two star earths..worked better...because you don't want large currents near small currents... |
|
|
| bobhayes |
Bas,
Every component which requires grounding is grounded by it's lead (or short wire) directly to the chassis, at the closest socket or transformer mounting nut.
The chassis layout is (from left to right):
Mains In
Mains Tx
Mains switch and fuse
Rectifier
Choke and HT caps
Speaker terminals
Output Tx's
2A3's
6SL7's
Volume and inputs
Its fairly conventional in order to eliminate such problems. |
|
|
| Bas Horneman |
Hi Bob,
A picture would say more than a thousand words.. :)
Because from your description...it does indeed sound as if that could be a problem...In Morgan Jones's book he describes how moving one wire 2 centimeter solved the hum..
Regards,
Bas |
|
|
| bobhayes |
Which wire?.. He,he,he!
Looks a mess, doesn't it? |
|
|
| SY |
| Thoroughly neater than anything I've ever wired up. |
|
|
| Bas Horneman |
That's pretty neet indeed...I can say the same as SY..that is neater than anything I've ever concocted..
How much ripple do you have on the B+ ?
Is your 6SL7 heater supply referenced to earth either via a real center tap or a pseudo center tapped (with resistors)? |
|
|
| bobhayes |
Don't mock me!
1.4mV ripple on the main B+ (Output Tx feeds).
Resistor network to bias 6SL7 heaters (76.5v).
So, at a glance, do you see any real flaws in the layout?
You get the general picture
Mains left, output middle, input right. |
|
|
| Bas Horneman |
He..he Bob ...not mocking you..seriously...:D
I build my stuff in wooden enclosures..that way one is ensured that where you want earth to be that is earth.
A tip by someone was also to have the transformers isolated from the topplate except.
At a glance I see no problems.. Are the RCA plugs isolated from the chassis? |
|
|
| Bas Horneman |
| Ok the biasing is good..but is there a center tap on the filament or did you simply ground one side of the filament supply...? |
|
|
| bobhayes |
Two resistors make up the divider network, lower resistor bypassed with 1uF capacitor to ground.
76.5v reference to one side of heater winding only.
As I originally said though, I can pull the 6SL7's and the hum remains the same so I don't think this is where the hum is being introduced.
Any other idea's? |
|
|
| Bas Horneman |
| quote: | | can pull the 6SL7's and the hum remains the same so I don't think this is where the hum is being introduced. |
Well I'm not saying it is...but if you pull the 6SL7 the heater wiring remains..
Lemme think..
O yes..RCA plugs isolated from chassis? |
|
|
| Sherman |
Bob,
You said your grounds were by wire to the closest screw or bolt connected to the chassis. I think I would do that a little differently. Keep in mind I'm a relative newbie having built only five or six tube amps and never a DHT amp.
So... here is my general idea of grounding.
1- Connect the mains earth wire to the chassis near the point where the mains enter the chassis.
2- Ground all the power components to a single point (not to the chassis).
3- Ground all the signal components to a different single point (not to the chassis).
4- Run separate heavy gauge (12 to 16 AWG- sorry, don't know the mm equivalent in millimeters) copper wires from each ground point back to the chassis ground. I tend to use solid core for this but have also used stranded.
5- Use isolated RCA jacks for the input and don't ground them to the chassis but to the 'signal ground point'.
6- Use shielded wire to connect the RCAs to the volume pot (or input selector if you have one) and from the volume pot to the input tube.
Ground loops can definitely be caused by the chassis. The voltage potential between two points on the chassis with all the components mounted and powered up can be significant. By not using random points all over the chassis for grounding you can reduce the loops. |
|
|
| kmtang |
I built my JE Lab 2A3 monoblocs with very minimal hum. It's less than 2mVp-p.
I had some buzz noise when the amp was connected to the Cable TV box (even with it switched off). The ground from the Cable TV box causing it to buzz or hum a lot. Finally, I had to isolate the ground wire from the Power Sockets in the 2A3 amps. The system ground will now be connected from the amps to the cable box, then to the cable wire's grounding.
Did you try to remove all the sources cables and see if it still hums???
Johnny |
|
|
| bobhayes |
Bas,
If the tubes are pulled, then there is no current flowing through the heater wires, and since the wires are twisted together, how could this influence the output?
Yes, the RCA's are connected to the chassis, not isolated, but again, how can a ground loop be caused when the rest of the signal path is broken by pulling the tubes?
Sherman,
In hindsight I should perhaps have arranged the grounding differently, but I'll try and isolate the problem first before rewiring the grounds, as this might not be the problem. Thanks for info though, as it will be useful if I have to rewire it.
What could cause hum on the speaker outputs with the input tubes AND rectifier pulled out?
Thanks for the input. |
|
|
| Bas Horneman |
| quote: | | If the tubes are pulled, then there is no current flowing through the heater wires, and since the wires are twisted together, how could this influence the output? |
There would still be a voltage present..?? I think...maybe injecting a little noise into your b+? I dunno..
Onto the RCA plugs..
The one 300B amp we built....was humming pretty bad until we isolated the rca plugs from the chassis.
How can you have humming with the rectifier pulled?.. a ground loop in your system could cause that. |
|
|
| bobhayes |
kmtang,
Yes, it still hums when there is nothing connected to the RCA's
Non of the sources I use have earth cables. The only earth comes through the RCA leads. |
|
|
| bobhayes |
I have found an AC signal (4.1v) between the earth cable and the neutral cable on my mains supply.
Could this cause any problem? |
|
|
| bobhayes |
BTW,
The RCA's are grounded to the chassis, but the screened cable to the volume pot is only grounded at the RCA end, so there's no loop there. |
|
|
| bobhayes |
Can't be the earth/neutral signal.
I just disconnected the mains earth lead from the chassis and it still hums.
Looks like I might have to move some grounds then! |
|
|
| stalker |
Hi,
I had a similar problem and also what i did was to check ground connections. In the end it turned out to be the power trafo that was coupling noise to my OT. Just had to move it away a few centimeters and the hum was gone. But your layout seems correct so i don't know if this is your problem.
Al. |
|
|
| JimW |
I have built a few amps using both a star ground technique and a heavy ground bus grounded at the input. I have been able to get hum down to less than 2mv (using ac fil) in every case, however a few things I have noticed that will help reduce hum:
1. Play with polarity of the ht ac leads (a Gordon Rankin tip) - usually one direction will show a little less hum.
2. Try using an external fil trans for 2a3 to see if you are coupling back through the main xformer.
3. The polarity of primary winding on the output trans. I had the wrong color scheme for transformer leads on one amp - I was having trouble with hum - wired the output trans correctly and the hum dropped to about one third of its previous level.
4. Try a different 2a3 (probably a long shot).
I think it is a good idea to put a resistor from center tap to each leg of the hum put (22-47 ohm) so that the humpot functions as a very fine adjustment, often the difference between noisey and quiet is a very small movement of the pot.
Good luck and enjoy your amps.
Jim |
|
|
| Sheldon |
| quote: | Originally posted by bobhayes
What could cause hum on the speaker outputs with the input tubes AND rectifier pulled out?
Thanks for the input. |
Hmm (no pun, just thinking), if the rectifier and driver tubes are pulled out, then the only current flowing in the system is the current for the heater supplies. This assumes that the caps have discharged. In this case, somehow there is coupling between the power tranny or wiring and output trannies? I would think that a ground issue that puts noise directly into the RCA would be 50hz, no?
Sheldon |
|
|
| tubelab.com |
There are two possible paths for hum in this case and I have seen both of them. One is a magnetic coupling between the output transformers, and the power transformer. If you look at the amp in the picture on my web site: http://www.tubelab.com/index.html
you will see what NOT to do. This amplifier always hummed, the hum was noticeable when the amp was switched on and died down as the output tubes warmed up. It didn't bother me on my 87db speakers, but when I took the amp to a friends house with 97db speakers, it was bad. I have since rotated the power transformer 90 degrees and the hum is gone. The hum was audible with ALL of the tubes removed from the amp, but got louder with the tubes in.
The 833A amp prototype had a pretty loud hum that appeared after the output tube warmed up, WITHOUT ANY other power supply turned on! Thats right, ONLY the filament transformer to the 833A is powered up and hummmmmmmm. Moving the output and filament transformers around had no effect. Breaking the plate circuit anywhere caused the hum to stop. The hum balance pot had no effect (until the rest of the amp was powered up). Breaking the grid path had little effect. Reversing the filament leads made a slight improvement. The hum appeared to be mostly 60 Hz until the whole amp was powered up, and then there was obviously a strong 120 Hz component, this varied with the hum balance pot, but in no case could the amp be made quiet. The prototype has since been dismantled, but I decided that the final design would need DC filaments, all 10 Amps worth. It would seem that sufficient current was flowing through the tube with only filament voltage present. I later confirmed that the plate will go about 6 volts negative with respect to the filament on filament voltage alone (big filament 10 volts 10 amps).
http://www.tubelab.com/833SE.htm
Listen carefully to your amp, does the hum appear as soon as the amp is switched on, and change slightly as the tubes warm up (rectifier and driver tubes removed)? Does it hum at all with the output tubes removed? If so then I would suspect magnetic coupling.
Does the hum appear as the output tubes warm up, silent on initial switch on (rectifier and driver tubes removed)? 2A3's warm up quick, 2 or 3 seconds. If so is there any other source of voltage in the amp? If there is no other source of voltage, try to find a simple way of breaking the plate circuit path. If the hum goes away, it MUST be a current that is flowing through the tube. How do you fix this? I don't exactly know. I have used DC filaments. Just exactly why some amps and some tubes exhibit this problem worse than others, I don't know. |
|
|
| Miles Prower |
| quote: |
I've been listening to my 2A3 amp for about a year, and love it. But I recently plugged my TV into it to watch a film, and the hum is unbearable for movies because of all the silence in films.
|
If the amp sounded OK before connecting it to the TV, then I would strongly suspect magnetic coupling from the vertical deflection coils to the amp. I once made the mistake of trying to record audio off a TV set with a microphone, and got some nasty background noise from the vertical deflection coils to the mic coil.
Yuck! :dead: |
|
|
| bobhayes |
Thanks for all the replies.
I've tried a few different things and have managed to get the hum down to 7.5mV, which seems to have made an improvement.
I swapped primary leads on the output tx's and the mains H.T. secondaries,
moved the main earthing point to the C.T. position of the H.T. secondaries, and re-arranged the heater wires. Not sure which of these made the greatest improvement (if any) but I think I'll have to live with the 8mV.
Moving the choke around made no difference and the mains Tx is about 6 inches away from the output Tx's, and I can't move it easily so I won't.
The mains Tx is isolated from the chassis with rubber feet to stop hum being transmitted into the chassis as the Sovteks are quite microphonic.
I can live with the residual hum as long as I don't use it for AV!
Thanks again everyone... Bob
P.s. At switch on the hum rises from zero, peaks after about one second then dies back to it's 7.5mV level. All the transformers and choke on this amp are C cores. Any further discussion welcomed! |
|
|
| Bas Horneman |
Hi Bob,
So far there have been a lot of suggestions..if possible it is probably easiest though to try to exclude one or 2 things...
First of all try DC on both the driver and the 2A3...? If that cures your hum...you know what is causing it. If not ...you have very likely excluded a probable cause.
And a further suggestion would be to remove the input rca from chassis contact! It's a very easy thing to do...connect the ground via a wire to any of your star earthing points.
Regards,
Bas |
|
|
|
