After replying to a posting elsewhere on filament power supplies, etc., I went and measured across the secondaries of my filament trafo. The center-tap seems exactly on center. I have the center-tap of the 6.3VAC Hammond referenced to 1/4B+ (about 79V) using a resistor in series with a cap from B+ to GND, with the center-tap in the middle. This seems pretty standard.
However, what I did notice is that I connected the divider to GND on at the GND lug of the second 45uF cap in my PSU. My PSU is: diode bridge; 0.68uF; 10H; 45uF; 10H; 45uF.
So, the 1/4B+ voltage divider works, but should it be grounded directly to chassis GND? Maybe to the Aikido PCB star-earth (which is lifted by 20R above chassis/earth)? Could it be that somehow noise gets injected via this divider?
While B+ is off and the heaters are warming, the 6.3VAC trafo is dead quiet, when B+ is switched on, it begins to buzz very, very slightly.
Any ideas?
Charlie
However, what I did notice is that I connected the divider to GND on at the GND lug of the second 45uF cap in my PSU. My PSU is: diode bridge; 0.68uF; 10H; 45uF; 10H; 45uF.
So, the 1/4B+ voltage divider works, but should it be grounded directly to chassis GND? Maybe to the Aikido PCB star-earth (which is lifted by 20R above chassis/earth)? Could it be that somehow noise gets injected via this divider?
While B+ is off and the heaters are warming, the 6.3VAC trafo is dead quiet, when B+ is switched on, it begins to buzz very, very slightly.
Any ideas?
Charlie
If I understand your circuit, the cap (-) is the same as the (-) supply to the preamp board (star ground). Since the divider is drawing current from the supply, that's where it should go - the resistor that isolates the star ground should not carry current. The buzzing is probably due to distortion of the AC waveform by the B+ supply - it draws a pulsed current, containing harmonics of line frequency.
Tom,
Again this leads to the question of quality of my B+, which I thought was good (but I don't have a scope to use). I'll move the GND of the divider to the (-) of the final cap in case this works better, although I doubt it. Wouldn't the cap on the divider filter these line harmonics? Maybe I need a higher valued cap here.
My plan had been to transition to the "new" PSU this weekend, although I just noticed that my caps are onyl rated for 370VAC. I'll need to have another look at PSUDII to see whath voltage they may actually recieve. After rectification, B+ should be around 340V in the new PSU, so the caps may see quite a bit higher. Having said that, I seem to remember being told that AC rated caps could handle greater DC voltages than their AC ratings.
If my main PSU caps are not large enough for voltage tolerance, I'll postpone the new PSU until I build a new chassis. In the meantime, I may try to tame the mains voltage that the trafos actually see. They are rated at 115V, but typically see 124V. A couple of flameproof 5W resistors in series with the primaries should drop voltage down to something that they can handle better. It is also a cheap ad-hoc fix.
Charlie
Again this leads to the question of quality of my B+, which I thought was good (but I don't have a scope to use). I'll move the GND of the divider to the (-) of the final cap in case this works better, although I doubt it. Wouldn't the cap on the divider filter these line harmonics? Maybe I need a higher valued cap here.
My plan had been to transition to the "new" PSU this weekend, although I just noticed that my caps are onyl rated for 370VAC. I'll need to have another look at PSUDII to see whath voltage they may actually recieve. After rectification, B+ should be around 340V in the new PSU, so the caps may see quite a bit higher. Having said that, I seem to remember being told that AC rated caps could handle greater DC voltages than their AC ratings.
If my main PSU caps are not large enough for voltage tolerance, I'll postpone the new PSU until I build a new chassis. In the meantime, I may try to tame the mains voltage that the trafos actually see. They are rated at 115V, but typically see 124V. A couple of flameproof 5W resistors in series with the primaries should drop voltage down to something that they can handle better. It is also a cheap ad-hoc fix.
Charlie
370V AC caps are typically good for 1000VDC, though specs vary.
I was referring to line harmonics entering through the PRIMARY of the filament transformer. There will also be some pulsed current flowing in the secondary, but it's small compared to heater current... probably not significant, and unavoidable. The stiffer your B+ supply is, the more harmonic current it will draw (choke input would be better in this respect).
If your secondary voltage is high on the heater transformer, a resistor can drop it. If your B+ is high, decrease the first cap... will reduce harmonic current along with voltage. I wouldn't worry about voltage rating of primaries unless the transformer is close to saturation (unloaded current greater than 10% or so of expected full-load current).
One more possibility... the buzzing might be due to magnetic coupling between the two transformers - are they close together, windings on same axis?
I was referring to line harmonics entering through the PRIMARY of the filament transformer. There will also be some pulsed current flowing in the secondary, but it's small compared to heater current... probably not significant, and unavoidable. The stiffer your B+ supply is, the more harmonic current it will draw (choke input would be better in this respect).
If your secondary voltage is high on the heater transformer, a resistor can drop it. If your B+ is high, decrease the first cap... will reduce harmonic current along with voltage. I wouldn't worry about voltage rating of primaries unless the transformer is close to saturation (unloaded current greater than 10% or so of expected full-load current).
One more possibility... the buzzing might be due to magnetic coupling between the two transformers - are they close together, windings on same axis?
Tom,
I am using motor-run caps for the B+ although I can replace these with larger voltage caps if neccessary, although I reckon that they ought to be able to tolerate a higher voltage.
My transformers are mounted at right angles to each other, and they have around 1.5 inches separating them. I am also using a steel top plate for my Aikido (to which the hardware is mounted), when I rebuild, I'll use aluminum.
Anyway, once I look at PSUDII, I'll have a good idea about whether to install the new PSU. If my caps will likely see more than around 450V, I'll probably opt for caution and obtain higher specced caps.
Charlie
I am using motor-run caps for the B+ although I can replace these with larger voltage caps if neccessary, although I reckon that they ought to be able to tolerate a higher voltage.
My transformers are mounted at right angles to each other, and they have around 1.5 inches separating them. I am also using a steel top plate for my Aikido (to which the hardware is mounted), when I rebuild, I'll use aluminum.
Anyway, once I look at PSUDII, I'll have a good idea about whether to install the new PSU. If my caps will likely see more than around 450V, I'll probably opt for caution and obtain higher specced caps.
Charlie
Hi Charlie
Since I didn't receive several of your emails and haven't been following this thread I'm somewhat behind on this issue. Forgive me if I repeat what others have said or ask about already settled items. First, lets concentrate our efforts. As I said I'm way behind but it sounds like your going in several directions at once. Your power supply caps are perfectly suited for your application and about as high quality as you can use. Its not clear to me if you have a mechanical issue (put your ear to the transformers and chokes) or an electrical issue (put your ear to the speaker woofer). Reducing AC voltage or mounting the transformer with rubber grommets MIGHT reduce transformer buzzing which is a mechanical issue. Ground wire rerouting MIGHT effect hum. My small brain is struggling with your ground system. I have two grounds. One is chassis ground connected to the electrical wall plug ground. The second ground is every other ground point. The two grounds are connected to each other by ying/yang diodes. No hum! filament is raised by running a 300K resister string to the filament and grounding that same point thru a 100K resister and 10mf cap to the "all other" ground. I think that is what you do too. If it is, thats probably not your problem.
Since I didn't receive several of your emails and haven't been following this thread I'm somewhat behind on this issue. Forgive me if I repeat what others have said or ask about already settled items. First, lets concentrate our efforts. As I said I'm way behind but it sounds like your going in several directions at once. Your power supply caps are perfectly suited for your application and about as high quality as you can use. Its not clear to me if you have a mechanical issue (put your ear to the transformers and chokes) or an electrical issue (put your ear to the speaker woofer). Reducing AC voltage or mounting the transformer with rubber grommets MIGHT reduce transformer buzzing which is a mechanical issue. Ground wire rerouting MIGHT effect hum. My small brain is struggling with your ground system. I have two grounds. One is chassis ground connected to the electrical wall plug ground. The second ground is every other ground point. The two grounds are connected to each other by ying/yang diodes. No hum! filament is raised by running a 300K resister string to the filament and grounding that same point thru a 100K resister and 10mf cap to the "all other" ground. I think that is what you do too. If it is, thats probably not your problem.
Bruce,
I have been trying to figure out what to do myself. The hum is not serious, it is not audible with any level of music playing, but I know it is there.
My ground system is similar to yours (I believe). I have chassis GND to which my top plate is connected. I just checked and this is what I have: chassis GND connected to IEC earth. To the Chassis GND bolt, I have case of rubber grommit mounted B+ trafo and my Circuit GND, which includes the PSU GND (joins the PCB at its star-earth). A wire goes from PCB star-earth to chassis GND bolt via a yin-yan diode (actually a 25A bridge rectifier) / cap / resistor network (aka Rod Elliot's website). There is a lift on this ground of 20R.
What I do notice is this: when I switch on the preamp and my filaments are warming-up (ie. NO B+), there is no humming. When the B+ switches on, both B+ and 6.3VAC trafos begin to buzz. The B+ trafo only buzzes slightly and is almost inaudible even with my ear touching it. However, the 6.3VAC trafo does does have a clear buzz when I put my ear in contact with it. NOTE: it does not buzz until B+ is switched on, which has led me to believe that my 1/4B+ reference for the center-tap of the 6.3 is injecting something that is making it hum.
I can remount the 6.3 on grommits, and I do have some 20R 5W resistors to install on the 6.3VAC trafo's primary to reduce the 124 mains voltage to something more like 118VAC.
I also paid a visit to our favorite tube store (AES) and picked-up a 275-0-275 trafo, which I may switch out and go for tube rectification. I already have an octal socket on the Aikido - used by my Amperex delay tube, so a little rewiring should allow me to have around 340VDC via full-wave tube rectification. Part of me is nervous about trying this, athough there is minimal surgery involved - basically remove 125-0-125 trafo, pop out the Amperex, rewire its socket, wire to 0.68uF first cap of filter/choke network and Bingo! Except that this time I'll need to wire the center-tap of the trafo's 5V secondary to GND. I assume that the center-tap (0V) of the 275-0-275 gets left unconnected to anything.
Anyway, if the tube recitifcation works, I'll be able to remove the Amperex delay and Omron relay from my list of possible noise sources.
Regards,
Charlie
I have been trying to figure out what to do myself. The hum is not serious, it is not audible with any level of music playing, but I know it is there.
My ground system is similar to yours (I believe). I have chassis GND to which my top plate is connected. I just checked and this is what I have: chassis GND connected to IEC earth. To the Chassis GND bolt, I have case of rubber grommit mounted B+ trafo and my Circuit GND, which includes the PSU GND (joins the PCB at its star-earth). A wire goes from PCB star-earth to chassis GND bolt via a yin-yan diode (actually a 25A bridge rectifier) / cap / resistor network (aka Rod Elliot's website). There is a lift on this ground of 20R.
What I do notice is this: when I switch on the preamp and my filaments are warming-up (ie. NO B+), there is no humming. When the B+ switches on, both B+ and 6.3VAC trafos begin to buzz. The B+ trafo only buzzes slightly and is almost inaudible even with my ear touching it. However, the 6.3VAC trafo does does have a clear buzz when I put my ear in contact with it. NOTE: it does not buzz until B+ is switched on, which has led me to believe that my 1/4B+ reference for the center-tap of the 6.3 is injecting something that is making it hum.
I can remount the 6.3 on grommits, and I do have some 20R 5W resistors to install on the 6.3VAC trafo's primary to reduce the 124 mains voltage to something more like 118VAC.
I also paid a visit to our favorite tube store (AES) and picked-up a 275-0-275 trafo, which I may switch out and go for tube rectification. I already have an octal socket on the Aikido - used by my Amperex delay tube, so a little rewiring should allow me to have around 340VDC via full-wave tube rectification. Part of me is nervous about trying this, athough there is minimal surgery involved - basically remove 125-0-125 trafo, pop out the Amperex, rewire its socket, wire to 0.68uF first cap of filter/choke network and Bingo! Except that this time I'll need to wire the center-tap of the trafo's 5V secondary to GND. I assume that the center-tap (0V) of the 275-0-275 gets left unconnected to anything.
Anyway, if the tube recitifcation works, I'll be able to remove the Amperex delay and Omron relay from my list of possible noise sources.
Regards,
Charlie
OK. Now I feel we are getting somewhere. The problem is with transformer/choke buzz. Seems logical to do the quick and/or easy stuff relating to that area first. I'm thinking the resister AC voltage reduction fits the bill. I can't see how the heater lift process could be at fault but its real easy to leave the power amp off, unsolder the splitter resisters and turn the pre on. Does the transformer still buzz? Two quick and easy things to try.
Bruce,
I'll try that first chance I get, maybe tomorrow before work. As I say, I have two 20R resistors rated at 5W, I'll put one of these on one of the 6.3VAC primaries and see what the voltage drops to.
Incidentally, I can put my ear of either choke and cannot hear the slightest buzz. Would a ringing choke not make at least a little buzz? Or maybe a ringing choke may simply pass the noise elsewhere.
Charlie
I'll try that first chance I get, maybe tomorrow before work. As I say, I have two 20R resistors rated at 5W, I'll put one of these on one of the 6.3VAC primaries and see what the voltage drops to.
Incidentally, I can put my ear of either choke and cannot hear the slightest buzz. Would a ringing choke not make at least a little buzz? Or maybe a ringing choke may simply pass the noise elsewhere.
Charlie
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.