Is it a 60Hz (50Hz in Europe) or 120Hz (100Hz in Europe) hum ?
Two Types of Tube Amplifier Hum and How to Determine the Source - YouTube
Two Types of Tube Amplifier Hum and How to Determine the Source - YouTube
Update. Just found a silly wiring mistake made in the process of making the other changes. Hum has gone from deafening to quite quiet. Seems like we’re on the right track
So, for the benefit of all the folks who contributed, and those who may read this thread later, can you summarize all the changes you made to get rid of the hum?
Certainly. I'm not actually free of hum yet, but it's gone from being an overwhelming sound audible through a closed door to being an annoying hum that's masked by moderate volume music.
- The thing that seemed to make the main difference was adding the virtual centre tap to the 6SN7 heater circuit.
- I also removed the 200H chokes and replaced them with resistors (I hadn't noticed the discrepancy between the two schematics). I've tried with and without them, and can't discern any difference. Once I've got rid of the rest of the hum I'll try them in and out again, but expect I'll be able to do without them.
- Changing from the paired resistors on the 300B filament to a hum pot made a small but important difference.
- Attaching the safety earth to signal earth induces dirty-sounding (not pure tone) noise. It's currently disconnected, but this doesn't seem right in the long term.
-I've experimented with moving the DC supplies further from the PT, but this didn't make any difference.
After doing all of the above, I forgot to re-attach one ground wire, which made an awful noise, but was an easy fix.
I'm not quite sure where to go next. My star ground currently attaches to the negative leg of the rectifier cap/CT, which I've read can be a problem. I'm thinking of changing that to have the rectifier ground attach only the the ground of the second (double) decoupling cap, then make that the centre of the star. Does that seem sensible?
Also still wondering if that rattling sound from the tubes while powered on is normal.
- The thing that seemed to make the main difference was adding the virtual centre tap to the 6SN7 heater circuit.
- I also removed the 200H chokes and replaced them with resistors (I hadn't noticed the discrepancy between the two schematics). I've tried with and without them, and can't discern any difference. Once I've got rid of the rest of the hum I'll try them in and out again, but expect I'll be able to do without them.
- Changing from the paired resistors on the 300B filament to a hum pot made a small but important difference.
- Attaching the safety earth to signal earth induces dirty-sounding (not pure tone) noise. It's currently disconnected, but this doesn't seem right in the long term.
-I've experimented with moving the DC supplies further from the PT, but this didn't make any difference.
After doing all of the above, I forgot to re-attach one ground wire, which made an awful noise, but was an easy fix.
I'm not quite sure where to go next. My star ground currently attaches to the negative leg of the rectifier cap/CT, which I've read can be a problem. I'm thinking of changing that to have the rectifier ground attach only the the ground of the second (double) decoupling cap, then make that the centre of the star. Does that seem sensible?
Also still wondering if that rattling sound from the tubes while powered on is normal.
By attaching signal and earth ground you likely created a ground loop. Read section 15.10 in this doc
http://www.valvewizard.co.uk/Grounding.pdf
and try implementing a simple ground loop breaker:
An externally hosted image should be here but it was not working when we last tested it.
This circuit is explained by Pete Millett in the following article:
DCPP Amp
Last edited:
If I read your picture correctly: you have two white wires
running from North-West to South.
Fig 15.3 http://www.valvewizard.co.uk/Grounding.pdf
suggests chassis grounding at the ingress.
running from North-West to South.
Fig 15.3 http://www.valvewizard.co.uk/Grounding.pdf
suggests chassis grounding at the ingress.
After reworked the PS, I realized that I don't need 22 ohm resistor between signal ground and chassis.
My 6.3v AC has center tap so I decided to try again with center tap and there is hum but the weird thing is only 1 channel got hum even thought this AC uses for both channels of 2 6xn7 and 1 12ax7. So I removed the center tap and put back the 2 of 100 ohm on each legs of 12ax7 then connect them to ground.
My 6.3v AC has center tap so I decided to try again with center tap and there is hum but the weird thing is only 1 channel got hum even thought this AC uses for both channels of 2 6xn7 and 1 12ax7. So I removed the center tap and put back the 2 of 100 ohm on each legs of 12ax7 then connect them to ground.
Hi again
With the advice above and some more research, I've managed to decrease the hum further, but have got stuck again. I've changed the grounding to have the centre of the star at the input, and have experimented with some other ground layouts without significant improvement. This thread understanding star grounding gave me some great pointers for getting the grounding right.
I found this excellent guide to troubleshooting hum by Vinylsavor, VinylSavor: Trouble Shooting Hum, and by following it through, know that the hum comes from the output stage. It doesn't even change with the 6SN7 removed.
The ripple current entering the OPT is 4.6mV, and the ripple on the DC supply to the 300B heater is 6mV on one and 4.6 mV on the other. The hum at the speaker is 4 mV.
While I've now narrowed down the part of the circuit responsible for the hum, I'm not sure how to narrow it down further. Is the power supply ripple low enough to say that's not the problem? Is there somewhere else I should be looking?
David
With the advice above and some more research, I've managed to decrease the hum further, but have got stuck again. I've changed the grounding to have the centre of the star at the input, and have experimented with some other ground layouts without significant improvement. This thread understanding star grounding gave me some great pointers for getting the grounding right.
I found this excellent guide to troubleshooting hum by Vinylsavor, VinylSavor: Trouble Shooting Hum, and by following it through, know that the hum comes from the output stage. It doesn't even change with the 6SN7 removed.
The ripple current entering the OPT is 4.6mV, and the ripple on the DC supply to the 300B heater is 6mV on one and 4.6 mV on the other. The hum at the speaker is 4 mV.
While I've now narrowed down the part of the circuit responsible for the hum, I'm not sure how to narrow it down further. Is the power supply ripple low enough to say that's not the problem? Is there somewhere else I should be looking?
David
David,
1. what's the make/model/sensitivity of your speakers?
2. the original build article indicated "the power supply filtering sections have been duplicated for each channel. While the power transformer (T2) is shared between both channels, each audio channel has separate rectifier, chokes and filter capacitors". This means four chokes, but in your design there are only two on the plate topside.
3. caps C6 and C7 should be quite physically large - did you replace these with the paralled caps under the chassis?
All in all, I think it's time for an updated schematic of the circuit, as built, with any value and design changes.
1. what's the make/model/sensitivity of your speakers?
2. the original build article indicated "the power supply filtering sections have been duplicated for each channel. While the power transformer (T2) is shared between both channels, each audio channel has separate rectifier, chokes and filter capacitors". This means four chokes, but in your design there are only two on the plate topside.
3. caps C6 and C7 should be quite physically large - did you replace these with the paralled caps under the chassis?
All in all, I think it's time for an updated schematic of the circuit, as built, with any value and design changes.
I got loud hum if connect ground house to signal ground but chassis is fine. After the hum is gone, I put back the ground house to the signal ground, it did hum again so it's one of the case on my amp.
At first I thought the star ground location is bad because near the B+ so I moved around but didn't change anything. The original hum was around around 40db
There may be another cause as well since i did all at the same time before fixing the hum. it was a high ripple with C filter so I changed to CRC filter. Also add 1uf630v(0.47uf is fine) on PS at each location of RC filter. The hum start lower until I removed the 6.3v center tap and ground house, then it's completely silence. I have 96 db sensitivity speakers.
Start without input to figure out the hum first. I made a mistake to connect a 6 foot rca input cable while trouble-shooing which also caused hum. You can see the image below to see where my GND at.
At first I thought the star ground location is bad because near the B+ so I moved around but didn't change anything. The original hum was around around 40db
There may be another cause as well since i did all at the same time before fixing the hum. it was a high ripple with C filter so I changed to CRC filter. Also add 1uf630v(0.47uf is fine) on PS at each location of RC filter. The hum start lower until I removed the 6.3v center tap and ground house, then it's completely silence. I have 96 db sensitivity speakers.
Start without input to figure out the hum first. I made a mistake to connect a 6 foot rca input cable while trouble-shooing which also caused hum.
You can see the image below to see where my GND at.
Read
Grounding Question
for earth-ground to chassis-ground connection.
It is not uncommon having a CL-80 thermistor between EG and CG.
I've seen a CL-80 & capacitor in parallel.
Grounding Question
for earth-ground to chassis-ground connection.
It is not uncommon having a CL-80 thermistor between EG and CG.
I've seen a CL-80 & capacitor in parallel.
fossa,
There are two separate amps being discussed in this thread:
The other builder, Alashikata, started a separate thread here:
Need schematic for this 300B to improve
There are two separate amps being discussed in this thread:
David added the hum pots which where not in the original picture/design.
The other builder, Alashikata, started a separate thread here:
Need schematic for this 300B to improve
Thanks for your thoughtful replies. Korneluk, to answer your points,
1. I’m using Visaton B200s, which are 96dB/W/m, but have also tried some AR18s’s at 86dB and the hum was still audible.
2. The 200H chokes (L1) are the two gold chokes inside the chassis. They’re currently out of the circuit and replaced with resistors, which didn’t seem to make any difference. This would fit with the observation that the hum seems to be coming from the output stage.
3. C6 and C7 are there. They’re electrolytic, so they’re not huge. C6 are the pair of burgundy ones at the bottom right and C7 are the other two burgundy pairs.
I’ve attached an up to date schematic.
Soulmerchant, good suggestion. Adding the hum pots back in did help, but I’ve already done it.
1. I’m using Visaton B200s, which are 96dB/W/m, but have also tried some AR18s’s at 86dB and the hum was still audible.
2. The 200H chokes (L1) are the two gold chokes inside the chassis. They’re currently out of the circuit and replaced with resistors, which didn’t seem to make any difference. This would fit with the observation that the hum seems to be coming from the output stage.
3. C6 and C7 are there. They’re electrolytic, so they’re not huge. C6 are the pair of burgundy ones at the bottom right and C7 are the other two burgundy pairs.
I’ve attached an up to date schematic.
Soulmerchant, good suggestion. Adding the hum pots back in did help, but I’ve already done it.
Attachments
One more test:
remove the 300B and measure the ripple voltages again. if the B+ is clean, or the ripple is less than before, then the hum is being introduced by the DC heater supply.
then, you can try measuring the ripple at the input pin to the voltage regulator and see how it compares to the output.
also, are you using the LT1085 regulator? can you post the exact part number? some of the regulators in the family, i.e. LT1083, etc., have different ripple rejection specs, so it may be worthwhile looking at the data sheets. Here's a good description and simulation of a similar supply:
Regulated DC filament supply
Last edited:
Hi Korneluk. Sorry about the slow reply: I've been away for a few days. I've just measured the B+ with the 300B removed, and the ripple is 4.8mV. It would seem from my reading that that's not enough to account for 3mV at the speakers, is that right?
The ripple across the DC to the 300B heaters is 4mV both with the tube in and removed, which again seems too low to be the culprit. I'm using the LT1085 CT regulator.
I found one mistake, where when I temporarily added the hum pot, I'd left the paired 47R resistors attached in parallel with the hum pot (but not connected to the cathode resistor). Properly removing them dropped the hum by about 2mV at the speakers. This has made me wonder whether adding extra resistance to each leg of the hum pot might be helpful. Would doing that have any untoward consequences?
The ripple across the DC to the 300B heaters is 4mV both with the tube in and removed, which again seems too low to be the culprit. I'm using the LT1085 CT regulator.
I found one mistake, where when I temporarily added the hum pot, I'd left the paired 47R resistors attached in parallel with the hum pot (but not connected to the cathode resistor). Properly removing them dropped the hum by about 2mV at the speakers. This has made me wonder whether adding extra resistance to each leg of the hum pot might be helpful. Would doing that have any untoward consequences?