Greetings from FixitLand!
I definitely agree with the idea of replacing the paired-in-series filter caps with single, properly-rated units. The Kendeil caps you specified above are 680-uF 400 V units. A bit spendy in my estimation, but be that as it may. I gather you propose to parallel them with the 10 uF 400 V Solen units? Not sure what the benefit of that would be. But any hum that's still going on with those new parts installed would have to be either grounding-related or caused by supersonic oscillation.
You said you measured 10.5-11 VAC at the EL84 anodes. What is the AC value at the output transformer's primary center-tap (B+)? If the level is about the same, it's definitely a power-supply issue. But if it's much lower, then what you measured at the anodes is a signal being amplified (oscillation, noise pickup). Remember than when you grounded the EL84 grids, the hum was silenced.
Take care,
--
J. E. Knox "The Victor Freak"
I did measurements, and i have some news. The hum that goes to the speaker, is 1.6Volts. I decided to check the psu again, and realized: there isn't enough bypass capacitors. I disassembed the PSU, made a Graetz rectifier on it, and added more capacitors. So the set is now: 470uF 200V capacitors in series, then a 120Ohm 5W resistor, then 330uF 200V capacitor in series, plus another 470uF 200V in series, plus another 100uF 400V...) So, i have 685uF now. Turned it on, measured the hum; it's 0.4Volts. I decided to buy new, good quality bypass capacitors. Solen and Kendeil. What do you think?
I definitely agree with the idea of replacing the paired-in-series filter caps with single, properly-rated units. The Kendeil caps you specified above are 680-uF 400 V units. A bit spendy in my estimation, but be that as it may. I gather you propose to parallel them with the 10 uF 400 V Solen units? Not sure what the benefit of that would be. But any hum that's still going on with those new parts installed would have to be either grounding-related or caused by supersonic oscillation.
You said you measured 10.5-11 VAC at the EL84 anodes. What is the AC value at the output transformer's primary center-tap (B+)? If the level is about the same, it's definitely a power-supply issue. But if it's much lower, then what you measured at the anodes is a signal being amplified (oscillation, noise pickup). Remember than when you grounded the EL84 grids, the hum was silenced.
Take care,
--
J. E. Knox "The Victor Freak"
Greetings from FixitLand!
He added 100-ohm screen resistors when he rebuilt the circuit on a piece of fiberboard. They should have been mounted closer to the socket though; looks like about an inch of wire between resistor body and tube socket.
Take care,
--
J. E. Knox "The Victor Freak"
He added 100-ohm screen resistors when he rebuilt the circuit on a piece of fiberboard. They should have been mounted closer to the socket though; looks like about an inch of wire between resistor body and tube socket.
Take care,
--
J. E. Knox "The Victor Freak"
Greetings from FixitLand!
"... I disassembed the PSU, made a Graetz rectifier on it ,..."
Hold it a sec...According to the PSU drawing you first posted, you are using an Edcor 180-0-180V (250 mA) power transformer; correct? That drawing shows a typical full-wave rectifier and C-R-C-R-C filter that should provide the 240 VDC and 200 VDC outputs as shown. But if you change the rectifier to a Graetz (i.e. bridge rectifier), connect it to the entire secondary and lift the center-tap, you will be rectifying **360 VAC** for a DC result of over 500 VDC! There's no benefit (and much hazard!) to doing that. And if you left the transformer center-tap grounded, plus tried to ground the B- leg of the bridge rectifier, two of the diodes would be blown up right quick. You *can* connect the bridge to half of the power tranny secondary, but that's just wasting half the tranny for no benefit. The simple two-diode full-wave rectifier is just fine.
Take care,
--
J. E. Knox "The Victor Freak"
"... I disassembed the PSU, made a Graetz rectifier on it ,..."
Hold it a sec...According to the PSU drawing you first posted, you are using an Edcor 180-0-180V (250 mA) power transformer; correct? That drawing shows a typical full-wave rectifier and C-R-C-R-C filter that should provide the 240 VDC and 200 VDC outputs as shown. But if you change the rectifier to a Graetz (i.e. bridge rectifier), connect it to the entire secondary and lift the center-tap, you will be rectifying **360 VAC** for a DC result of over 500 VDC! There's no benefit (and much hazard!) to doing that. And if you left the transformer center-tap grounded, plus tried to ground the B- leg of the bridge rectifier, two of the diodes would be blown up right quick. You *can* connect the bridge to half of the power tranny secondary, but that's just wasting half the tranny for no benefit. The simple two-diode full-wave rectifier is just fine.
Take care,
--
J. E. Knox "The Victor Freak"
Last edited:
No, i have 250Volts DC, because my transformer has 190V-0V-190V. I put the graetz on the 0V-190V, not on the 190V-190V.
AC voltage is 1.1Volts with my original PSU, now it's 0.4Volts, but i still hear a buzz. I think the EL84 tubes amplify the buzz, because the 0.22uF capacitor, (that removes the B+ voltage from the EL84s) is behave as a piece of wire when AC voltage is applied to it (XC=1/2*pi*f*C). So, if the hum is 0.4Volts, and if the EL84s amplification is lets say 10 times, than the hum is going to be 4Volts. Is this possible?
Greetings from FixitLand!
"I put the graetz on the 0V-190V, not on the 190V-190V." OKeh; yes, that will work, but it loads only half the transformer's output. Using the standard two-diode full-wave bridge is the better solution and gives the same voltage output to the filters.
"AC voltage is 1.1Volts with my original PSU, now it's 0.4Volts, but i still hear a buzz. ..."
The 0.4 VAC is measured from the OPT center-tap (B+) to ground, correct? That's not great.
"...I think the EL84 tubes amplify the buzz, because the 0.22uF capacitor, (that removes the B+ voltage from the EL84s) is behave as a piece of wire when AC voltage is applied to it (XC=1/2*pi*f*C). So, if the hum is 0.4Volts, and if the EL84s amplification is lets say 10 times, than the hum is going to be 4Volts. Is this possible?"
The EL84s aren't amplifying hum on their plate supply (but more on that below); in fact, the plates are relatively insensitive to power-supply variations. They *are,* however, sensitive to what's on their screens. You might measure the AC present on each screen (from there to ground). I'm not sure it's gonna make much difference though. Recall that the buzz was gone when the EL84 grids were grounded. Now that the phase inverter is in operation, we should bypass its input to see if the noise is eliminated. Did you do that yet? I thought you did...You are using pins 6, 7 & 8 of the ECC82, correct? Temporarily connect a 0.1-uF 400 V capacitor from pin 7 of the ECC82 to ground, then power up (no input to the EF86, or just pull it first). Should be quiet. You can touch a screwdriver to pin 7 of the ECC82 (don't touch the metal, as around 70-85 V should be present!), and some noise should result. That's good.
Which 0.22-uF capacitor do you mean? I don't see that value on the schematic. Perhaps you refer to C4 or C5, which do in fact block B+ from the EL84 grids. Yes, those capacitors are effectively "wires" to the AC signal. If there is indeed a hum signal there, the EL84s will amplify it. Now, the schematic shows that the phase-inverter's B+ supply is the same as that of the EL84s. One thing we have not tried is decoupling the phase-inverter's B+. If you can tack a 100-ohm 2-watt resistor (much like the units you used as EL84 screen resistors) between the B+ line and the top of R8 (the P.I.'s anode resistor), then add a 10-uF 400-V electrolytic from the top of R8 to ground, that should sufficiently decouple the P.I. without seriously lowering its supply voltage, and should reduce the buzz.
When you say "buzz" -- I wish you had a scope so we could see this noise signal. You mentioned 50 and/or 100 Hz, which is certainly mains-related...but that is usually called "hum" rather than "buzz." Buzz suggests there's other components to the noise. Is the noise "smooth" in character or "harsh"? And I just had another thought: Is there a light dimmer, fluorescent lamp or other electrical-noise-generating device in the vicinity? They can drive you nuts troubleshooting noise issues. Since this set is not built on a metal chassis, there's no shielding against such interference. I've built several similar amps just tacked together on my bench -- no cardboard, no fiberboard, no nothin' but air! -- and had no major noise problems, except for the lights in my record room which are all on dimmers; turning off the lights killed the noise.
Take care,
--
J. E. Knox "The Victor Freak"
"I put the graetz on the 0V-190V, not on the 190V-190V." OKeh; yes, that will work, but it loads only half the transformer's output. Using the standard two-diode full-wave bridge is the better solution and gives the same voltage output to the filters.
"AC voltage is 1.1Volts with my original PSU, now it's 0.4Volts, but i still hear a buzz. ..."
The 0.4 VAC is measured from the OPT center-tap (B+) to ground, correct? That's not great.
"...I think the EL84 tubes amplify the buzz, because the 0.22uF capacitor, (that removes the B+ voltage from the EL84s) is behave as a piece of wire when AC voltage is applied to it (XC=1/2*pi*f*C). So, if the hum is 0.4Volts, and if the EL84s amplification is lets say 10 times, than the hum is going to be 4Volts. Is this possible?"
The EL84s aren't amplifying hum on their plate supply (but more on that below); in fact, the plates are relatively insensitive to power-supply variations. They *are,* however, sensitive to what's on their screens. You might measure the AC present on each screen (from there to ground). I'm not sure it's gonna make much difference though. Recall that the buzz was gone when the EL84 grids were grounded. Now that the phase inverter is in operation, we should bypass its input to see if the noise is eliminated. Did you do that yet? I thought you did...You are using pins 6, 7 & 8 of the ECC82, correct? Temporarily connect a 0.1-uF 400 V capacitor from pin 7 of the ECC82 to ground, then power up (no input to the EF86, or just pull it first). Should be quiet. You can touch a screwdriver to pin 7 of the ECC82 (don't touch the metal, as around 70-85 V should be present!), and some noise should result. That's good.
Which 0.22-uF capacitor do you mean? I don't see that value on the schematic. Perhaps you refer to C4 or C5, which do in fact block B+ from the EL84 grids. Yes, those capacitors are effectively "wires" to the AC signal. If there is indeed a hum signal there, the EL84s will amplify it. Now, the schematic shows that the phase-inverter's B+ supply is the same as that of the EL84s. One thing we have not tried is decoupling the phase-inverter's B+. If you can tack a 100-ohm 2-watt resistor (much like the units you used as EL84 screen resistors) between the B+ line and the top of R8 (the P.I.'s anode resistor), then add a 10-uF 400-V electrolytic from the top of R8 to ground, that should sufficiently decouple the P.I. without seriously lowering its supply voltage, and should reduce the buzz.
When you say "buzz" -- I wish you had a scope so we could see this noise signal. You mentioned 50 and/or 100 Hz, which is certainly mains-related...but that is usually called "hum" rather than "buzz." Buzz suggests there's other components to the noise. Is the noise "smooth" in character or "harsh"? And I just had another thought: Is there a light dimmer, fluorescent lamp or other electrical-noise-generating device in the vicinity? They can drive you nuts troubleshooting noise issues. Since this set is not built on a metal chassis, there's no shielding against such interference. I've built several similar amps just tacked together on my bench -- no cardboard, no fiberboard, no nothin' but air! -- and had no major noise problems, except for the lights in my record room which are all on dimmers; turning off the lights killed the noise.
Take care,
--
J. E. Knox "The Victor Freak"
"Which 0.22-uF capacitor do you mean? I don't see that value on the schematic. Perhaps you refer to C4 or C5, which do in fact block B+ from the EL84 grids. Yes, those capacitors are effectively "wires" to the AC signal. If there is indeed a hum signal there, the EL84s will amplify it."
I meant those capacitors. There is a hum, because it's connected to the B+ (and as i said, the B+ has 0.4Volts hum).
I meant those capacitors.
There is a hum, because it's connected to the B+ (and as i said, the B+ has 0.4Volts hum).It's not a great solution (i did it once), because it's going to weaken the ECC82s output signal. But, if i connect the capacitor parallel with the 100Ohm, it's going to reduce the hum on the 100Ohm res. (actually 1.1Volts*(100/22100), according to the voltage-divider law).
Last edited:
Greetings from FixitLand!
OKeh now. I think part of the issue is those caps in series in the PSU. Two 470 uF in series becomes 235 uF while two 330 uF in series become 165 uF. If anything, the two 330 uFs should be right after the rectifier and the two 470 uFs should come after the 120-ohm resistor. You also need to add 220K resistors (value not critical) across EACH cap. This equalizes the voltage drop across the capacitors. Ideally, rather than caps-in-series, single capacitors of 350-400 V rating each should be used.
Then you should try decoupling the phase inverter as I mentioned previously. Let me know if you need more coaching on that. It's not hard.
Glad to hear you are free of noise-creating devices in the area.
I will have you do some more measurements after you decouple the P.I. Hang in there.
(I also realize it's getting late there, at this writing 22:11 [or 10:11 PM]...Here in western Oregon, USA, it is 14:11 [2:11 PM].)
Take care,
--
J. E. Knox "The Victor Freak"
OKeh now. I think part of the issue is those caps in series in the PSU. Two 470 uF in series becomes 235 uF while two 330 uF in series become 165 uF. If anything, the two 330 uFs should be right after the rectifier and the two 470 uFs should come after the 120-ohm resistor. You also need to add 220K resistors (value not critical) across EACH cap. This equalizes the voltage drop across the capacitors. Ideally, rather than caps-in-series, single capacitors of 350-400 V rating each should be used.
Then you should try decoupling the phase inverter as I mentioned previously. Let me know if you need more coaching on that. It's not hard.
Glad to hear you are free of noise-creating devices in the area.
I will have you do some more measurements after you decouple the P.I. Hang in there.
(I also realize it's getting late there, at this writing 22:11 [or 10:11 PM]...Here in western Oregon, USA, it is 14:11 [2:11 PM].)
Take care,
--
J. E. Knox "The Victor Freak"
Alright, then i'm going to modify the PSU.
Yes, it's 10'o clock here, but hey, tomorrow isn't even a school-day.
Greetings from FixitLand!
Looks like I'm getting some of your earlier replies after I last posted...that's confusing, no doubt!
Hmmm, no change with the P.I. grid bypassed. Decoupling the P.I.'s B+ *should* not materially affect the drive to the EL84s, since the P.I. plus the EF86 stage should be drawing less than 10 mA total and the drop across a 100-ohm resistor at that amperage is only 1 volt, leaving 249 V B+ for the P.I. You would NOT connect the 10 uF 400 V capacitor *across* the 100-ohm resistor; it goes from the top of R8 to GROUND, and acts as a filter for the B+. How did you connect the 100-ohm resistor when you tried it previously? Just curious.
Take care,
--
J. E. Knox "The Victor Freak"
Looks like I'm getting some of your earlier replies after I last posted...that's confusing, no doubt!
Hmmm, no change with the P.I. grid bypassed. Decoupling the P.I.'s B+ *should* not materially affect the drive to the EL84s, since the P.I. plus the EF86 stage should be drawing less than 10 mA total and the drop across a 100-ohm resistor at that amperage is only 1 volt, leaving 249 V B+ for the P.I. You would NOT connect the 10 uF 400 V capacitor *across* the 100-ohm resistor; it goes from the top of R8 to GROUND, and acts as a filter for the B+. How did you connect the 100-ohm resistor when you tried it previously? Just curious.
Take care,
--
J. E. Knox "The Victor Freak"
I'll be 17 years old in december. (My parents wasn't happy when i said 250Volts btw...
)Just to be clear; when you say P.I., you mean phase inverter, and the ECC82 tube right? And R8 is the 22kOhm that's connected to the ECC82s anode right?
Hmm. I did exactly what you said, and the hum disappeared. I can only hear it about a 15-20centimeter distance. Thanks for everything! My amp is working perfectly now. Oh, almost forgot it. I bought new speakers today. It's the Videoton (hungarian electrotechnical company, closed at 1991, after collapsing of the USSR) Preludium B32. This two way system was the entry-level audiophile sound system. It's equipped with VIFA speakers, just like the Heybrook HB-1 (actually i think the engineers at Videoton copied the design of the Heybrook). About it's sound: detailed with very powerful bass, and rich highs. Right now i'm using it with a homemade Hiraga LeMonstre 8W Class A amp (with original SanKen 2SA747 and 2SC1116 transistors).
I did exactly what you said, and the hum disappeared. I can only hear it about a 15-20centimeter distance. Thanks for everything! My amp is working perfectly now. Oh, almost forgot it. I bought new speakers today. It's the Videoton (hungarian electrotechnical company, closed at 1991, after collapsing of the USSR) Preludium B32. This two way system was the entry-level audiophile sound system. It's equipped with VIFA speakers, just like the Heybrook HB-1 (actually i think the engineers at Videoton copied the design of the Heybrook). About it's sound: detailed with very powerful bass, and rich highs. Right now i'm using it with a homemade Hiraga LeMonstre 8W Class A amp (with original SanKen 2SA747 and 2SC1116 transistors).
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.