Just out of curiosity, what is your cathode voltage running?
EDIT: Just saw your reply. I'm not sure what effect the 0.1uF cathode bypass will have on the high frequency response. Impedance of the cap is going to be about 100 ohms at 15kHz, which might roll your amp off at the high end. But at least now you have a data point on the hiss.
EDIT: Just saw your reply. I'm not sure what effect the 0.1uF cathode bypass will have on the high frequency response. Impedance of the cap is going to be about 100 ohms at 15kHz, which might roll your amp off at the high end. But at least now you have a data point on the hiss.
Last edited:
The mission now is to identify noise sources. I will optimize parameters as soon as we figure out how to kill that hum.
The hum is pretty faint but there. It is audible in the headphones. It occurs when the input floats, is shorted, or a source is connected.
The grid to cathode voltage is about -1.4 volts (just measured). So cathode voltage is 1.4 volts.
The hum is pretty faint but there. It is audible in the headphones. It occurs when the input floats, is shorted, or a source is connected.
The grid to cathode voltage is about -1.4 volts (just measured). So cathode voltage is 1.4 volts.
Very good.
Well, the hum might be coming from the Heathkit supply. I've not fired mine up yet, but it's been accused of not being quite "audio-quality".
Maybe now that the big hum is gone, it might be time to try your real B+ supply to see if it's any better and optimize from there.
As far as the cathode rolloff, if it's a problem you might be able to float the cathode on a fixed resistor to isolate it from the bypassed current sink. I think that raising the cathode up from the current sink (and tying the grids back to the bottom of the added cathode resistor) can retain your bias of 1.4-ish volts and get that bypass out of the picture as far as the audio signal goes.
Sounds like great progress, kingneb. Great work!
~ Sam
Well, the hum might be coming from the Heathkit supply. I've not fired mine up yet, but it's been accused of not being quite "audio-quality".
Maybe now that the big hum is gone, it might be time to try your real B+ supply to see if it's any better and optimize from there.
As far as the cathode rolloff, if it's a problem you might be able to float the cathode on a fixed resistor to isolate it from the bypassed current sink. I think that raising the cathode up from the current sink (and tying the grids back to the bottom of the added cathode resistor) can retain your bias of 1.4-ish volts and get that bypass out of the picture as far as the audio signal goes.
Sounds like great progress, kingneb. Great work!
~ Sam
Here is an update along with a new schematic. The schematic is incomplete as this is a work in progress.
I:
- Replaced the noisy current sink with a resistor and split rail supply (+/- 300V). There is no hiss and the bandwidth starts rolling off above 300Khz (probably due to the op amp).
- The filament supply now powers the op amps.
- The final circuit will have a SPI controlled digital volume control with a touchscreen interface. The 7805 powers all that.
Schematic flaws:
- Bridge rectifiers are wrong (you get the point though).
- 100 ohm resistors in power supply are 68k.
Remaining issues:
- There is still a very faint hum only audible only through headphones.
I:
- Replaced the noisy current sink with a resistor and split rail supply (+/- 300V). There is no hiss and the bandwidth starts rolling off above 300Khz (probably due to the op amp).
- The filament supply now powers the op amps.
- The final circuit will have a SPI controlled digital volume control with a touchscreen interface. The 7805 powers all that.
Schematic flaws:
- Bridge rectifiers are wrong (you get the point though).
- 100 ohm resistors in power supply are 68k.
Remaining issues:
- There is still a very faint hum only audible only through headphones.
Attachments
Now that things have quieted down, I was able to make out that hum better.
It is the kind of hum that has a minor buzz superimposed upon it. Do you think that is a ground loop still? What is the best way you think to kill that?
When the unit is first switched on, the buzz-hum is pretty loud and it becomes faint as the tube warms up.
When the unit is switched off, the AM 1400 comes in briefly until the tube cools off and caps discharge.
It is the kind of hum that has a minor buzz superimposed upon it. Do you think that is a ground loop still? What is the best way you think to kill that?
When the unit is first switched on, the buzz-hum is pretty loud and it becomes faint as the tube warms up.
When the unit is switched off, the AM 1400 comes in briefly until the tube cools off and caps discharge.
Last edited:
The tube is a 7025. The buzz increases in volume when you put your hand near or around the tube. AM 1400 comes in too. Same happens when you touch the coupling cap lead, only its becomes very loud when that happens.
An old Heathkit preamp I have does the exact same thing when you put your hand in the vicinity of the electronics of the tone controls, tubes, or phono stage.
An old Heathkit preamp I have does the exact same thing when you put your hand in the vicinity of the electronics of the tone controls, tubes, or phono stage.
Hi kingneb,
It's possible you're seeing the limits of the protoboard approach. If the hum is faint but increases when you add some hand-capacitance, I'm thinking the grounds are probably tingling a bit. Not unexpected with the protoboards - the thin wires that connect grounds together are short, but still tiny inductors at the AM radio frequencies.
The hum itself might be coupling from the power transformer - have you tried re-orienting the power transformer to see if the hum gets louder or softer? That level of hum can even couple into the headphones directly, even though they're low impedance.
Once things are installed into a good chassis with solid grounds, I think your hum might go away. From your description, it's pretty low now so it sounds like it's not a purely conducted problem in the circuitry.
~ Sam
EDIT: BTW, I think I'd put a good electrolytic cap right on the output of the LM7812 just to be sure you aren't getting a bit of ripple there. Something in the 10's of uF - doesn't really have to be huge, just provide a path for any residual low-frequency stuff.
Last edited:
A while ago, I finished my previous preamp, hum free.
Now I am making an all tube line stage. It is having some grounding issues as well. The schematic is attached. The power supply is just a 15k resistor pi filter and two 22uF capacitors. The filament is a DC heater powered by a 7812.
What is weird is when the 12BH7 tube is removed from the unit, there is still a low level hum. It grows worse when the tube is added back. Definitely something to do with the ground. The hum is virgin 60 cycles, not 120 or excessive buzz.
I am testing it with a Dynaco MKIII that needs an earth ground hum isolator (on the preamp) when used with a known good preamp. Ground loops always form between earth ground so I have the hum-x isolator attached. Therefore, I feel I eliminated that problem.
The tubes I am working with are bench test tubes, and microphonic. I doubt that is contributing to the hum as the filament is being powered by a 7812 regulator.
I had the thing build in a prototype chassis and a breadboard. Both hum. I am testing on the breadboard now.
Any ideal way to route the grounds to stop that issue? It is a different issue I am having from the previous unit, as that one just needed an earth reference.
Now I am making an all tube line stage. It is having some grounding issues as well. The schematic is attached. The power supply is just a 15k resistor pi filter and two 22uF capacitors. The filament is a DC heater powered by a 7812.
What is weird is when the 12BH7 tube is removed from the unit, there is still a low level hum. It grows worse when the tube is added back. Definitely something to do with the ground. The hum is virgin 60 cycles, not 120 or excessive buzz.
I am testing it with a Dynaco MKIII that needs an earth ground hum isolator (on the preamp) when used with a known good preamp. Ground loops always form between earth ground so I have the hum-x isolator attached. Therefore, I feel I eliminated that problem.
The tubes I am working with are bench test tubes, and microphonic. I doubt that is contributing to the hum as the filament is being powered by a 7812 regulator.
I had the thing build in a prototype chassis and a breadboard. Both hum. I am testing on the breadboard now.
Any ideal way to route the grounds to stop that issue? It is a different issue I am having from the previous unit, as that one just needed an earth reference.
Attachments
I rebuilt the thing on the breadboard and now it does not hum without a tube in the preamp. Corrected that issue.
With a tube in the system, when power is removed, the 60 cycle noise immediately stops. No hum while the tube cools down.
I added a 20 ohm resistor divider and another cap to power the second stage. The grounds for each stage return to the minus terminal of the filter capacitor of each stage. All those connections merge at the star where earth and the center tap of the transformer meet. Same is true for the 7812 circuitry.
With a tube in the system, when power is removed, the 60 cycle noise immediately stops. No hum while the tube cools down.
I added a 20 ohm resistor divider and another cap to power the second stage. The grounds for each stage return to the minus terminal of the filter capacitor of each stage. All those connections merge at the star where earth and the center tap of the transformer meet. Same is true for the 7812 circuitry.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.