I've just DIY a GG pre-amp, and burnt my power amp and speakers. I found that there is a 0.1V AC all over the chassis.
I've check everything in the GG but still couldn't find the cause of the AC. In the GG, I use star grounding: all components' ground connect to a single point and from this point to the chassis. I was told not to connect the earth of the AC supply to the chassis to have better S/N. But than I got this AC problem and have to connect the earth to the chassis to, sort of, resolve it.
How do you experts handle the grounding and earth?
I've check everything in the GG but still couldn't find the cause of the AC. In the GG, I use star grounding: all components' ground connect to a single point and from this point to the chassis. I was told not to connect the earth of the AC supply to the chassis to have better S/N. But than I got this AC problem and have to connect the earth to the chassis to, sort of, resolve it.
How do you experts handle the grounding and earth?
What a shame about your power amp and speakers! I find that hard to believe, are you sure the schematic was correct?
In my opinion, the chassis of the amplifier or preamp should ALWAYS be connected to mains safety ground/earth. It makes no difference to the S/N ratio in my experience.
My grounding is dual or triple star- I have one star ground for the filaments to chassis, another star ground for the signal/input grounds to the chassis, and then another to the cathodes/output tubes then to the chassis.
All I can suggest is making sure that you have something plugged into the preamp, otherwise you might get hum- and also that your heaters are earthed at one end...
In my opinion, the chassis of the amplifier or preamp should ALWAYS be connected to mains safety ground/earth. It makes no difference to the S/N ratio in my experience.
My grounding is dual or triple star- I have one star ground for the filaments to chassis, another star ground for the signal/input grounds to the chassis, and then another to the cathodes/output tubes then to the chassis.
All I can suggest is making sure that you have something plugged into the preamp, otherwise you might get hum- and also that your heaters are earthed at one end...
It never fails, whenever someone starts a sentence with, "I was told..." or "A friend of mine told me..", what follows is certain to be wrong.
As Shifty says, ALWAYS connect your chassis to earth (safety) ground. No exceptions.
Your signal ground can connect to the chassis at one point (usually the input socket); if there's a ground loop, remove the SIGNAL ground connection from the chassis, not the safety ground. You can reference the signal ground to safety ground through a small (say, 47 ohm) resistor.
There are quite a few ways to do signal/power supply grounding; I use a combination of star and ground bus. Morgan Jones has a nice explanation in "Building Valve Amplifiers." Aiken Amps also has a good write-up, though more appropriate for guitar amps than hifi.
As Shifty says, ALWAYS connect your chassis to earth (safety) ground. No exceptions.
Your signal ground can connect to the chassis at one point (usually the input socket); if there's a ground loop, remove the SIGNAL ground connection from the chassis, not the safety ground. You can reference the signal ground to safety ground through a small (say, 47 ohm) resistor.
There are quite a few ways to do signal/power supply grounding; I use a combination of star and ground bus. Morgan Jones has a nice explanation in "Building Valve Amplifiers." Aiken Amps also has a good write-up, though more appropriate for guitar amps than hifi.
Thank you for all the advises. Ground to earth is a must in my future project. But with this, is it safe to use the preamp in my case? My power amp and speakers are gone and I am about to use the preamp on their replacement.
The project is a kit with proven records. So it was just my mistake that cause the problem. But the point is, I am not able to find out why there is a 0.1V AC on the chassis. I thing grounding to earth is just a work around rather than a solution since there is still AC flowing on the chassis to earth. It may still damage other components.
What do you think is the likely cause of the AC?
The project is a kit with proven records. So it was just my mistake that cause the problem. But the point is, I am not able to find out why there is a 0.1V AC on the chassis. I thing grounding to earth is just a work around rather than a solution since there is still AC flowing on the chassis to earth. It may still damage other components.
What do you think is the likely cause of the AC?
senderj,
As SY said, """It never fails, whenever someone starts a sentence with, "I was told..." or "A friend of mine told me..", what follows is certain to be wrong."""
This is especially true in this place we are from. The 0.1V AC can be some sort of current "leaked out" from the power transformer. It should not be the cause of the death of your power amp and speakers. Perhaps you can try to reverse the connection of the primary secion of the power transformer and see if it will decrease (or increase). Some people say if you connect the inner end of the primary winding of the power transformer to the "L" point of the main AC, you will get more of this current leaked out from the power transformer.
by the way, what multi meter did you used to get that 0.1V A/C? Do you think it is good enough to give such a small reading accurately?
As SY said, """It never fails, whenever someone starts a sentence with, "I was told..." or "A friend of mine told me..", what follows is certain to be wrong."""
This is especially true in this place we are from. The 0.1V AC can be some sort of current "leaked out" from the power transformer. It should not be the cause of the death of your power amp and speakers. Perhaps you can try to reverse the connection of the primary secion of the power transformer and see if it will decrease (or increase). Some people say if you connect the inner end of the primary winding of the power transformer to the "L" point of the main AC, you will get more of this current leaked out from the power transformer.
by the way, what multi meter did you used to get that 0.1V A/C? Do you think it is good enough to give such a small reading accurately?
I use an ordinary non-digital Sanwa multimeter. The accuracy of 0.1V doesn't matter. What matter is there is some AC on the chassis and it goes away when switching the preamp off. In fact, it was measured with one leg on the chassis and another one on the window frame, which I use as the "earth". So the absolute voltage may not be 0.1V. But as said, the figure doesn't matter.
Thanks for the suggestion on the transformer. I will try to see if it does behave as said. Thanks again.
Thanks for the suggestion on the transformer. I will try to see if it does behave as said. Thanks again.
DC at power on and off
I finally found out what killed my speaker. I posted here on measuring nil to minimal DC voltage at the output of the pre amp before. It was measured when the pre amp was stable, i.e. long time after power on. Now I found that the DC shoots up a short while after power on and off. It is like this:
Power on: DC at pre amp output (gnd + and signal -) rised from 0V to 16V in 6-7 seconds, then down to -V and back to 0V (or close to zero).
Power off: DC (same polarity) rised from 0V to 12V in 12-13 seconds, then again down to -V and back to 0V.
I know the order of powering on should be pre then power, and off should be power then pre, and I did this all the time. But I didn't know I should wait for such a long time before switching the next one. My power amp has a power on delay for just about 4 seconds, which is not enough for my DIY pre.
Is there anything I can do in the pre to shorten the time of the DC or lowering its voltage?
I finally found out what killed my speaker. I posted here on measuring nil to minimal DC voltage at the output of the pre amp before. It was measured when the pre amp was stable, i.e. long time after power on. Now I found that the DC shoots up a short while after power on and off. It is like this:
Power on: DC at pre amp output (gnd + and signal -) rised from 0V to 16V in 6-7 seconds, then down to -V and back to 0V (or close to zero).
Power off: DC (same polarity) rised from 0V to 12V in 12-13 seconds, then again down to -V and back to 0V.
I know the order of powering on should be pre then power, and off should be power then pre, and I did this all the time. But I didn't know I should wait for such a long time before switching the next one. My power amp has a power on delay for just about 4 seconds, which is not enough for my DIY pre.
Is there anything I can do in the pre to shorten the time of the DC or lowering its voltage?
DC that is varying is in fact AC, or more accurately in this case, DC with an AC component. The change in a slow to rise voltage on the preamp side will be seen through too large a value of cap.
What is the input impedance of your power amp? If it is as high as 1Meg, a good cap size would be 80nF (0.08uF) but could go as small as 8nF. If your input impedance is as low as 10k, a good size would be 8uF but you could go as low as 0.8uF
What is the input impedance of your power amp? If it is as high as 1Meg, a good cap size would be 80nF (0.08uF) but could go as small as 8nF. If your input impedance is as low as 10k, a good size would be 8uF but you could go as low as 0.8uF
Hi senderj,
Why not use a signal relay to short the output of your preamp to ground until it has settled down? Make it so it shorts as soon as you turn off the power. Use the normally closed contacts so the outputs are shorted when there is no power.
No more clicks or bangs.
Ground the chassis now! It's not really safe to use otherwise. You could use a 10 R resistor to the circuit ground, your AC ground is hard (directly connected).
-Chris
Why not use a signal relay to short the output of your preamp to ground until it has settled down? Make it so it shorts as soon as you turn off the power. Use the normally closed contacts so the outputs are shorted when there is no power.
No more clicks or bangs.
Ground the chassis now! It's not really safe to use otherwise. You could use a 10 R resistor to the circuit ground, your AC ground is hard (directly connected).
-Chris
Amplifiers include input impedance as a standard spec, maybe you could find a spec sheet.
Failing that, try measuring with a multimeter. There is a good chance there will be an input coupling capacitor and if there is, and you measure at the RCA socket, you will just measure an open circuit. If you open the case you can then put one test lead just after the first capacitor (you'll need to trace it, ignore the disc caps, it will be film or electro), and one lead on the RCA socket outer terminal.
This will give a close practical approximation in most cases. It is better to get the spec from the manufacturer.
Failing that, just use a small value cap, and measure or listen for frequency response roll off. No harm really.
Failing that, try measuring with a multimeter. There is a good chance there will be an input coupling capacitor and if there is, and you measure at the RCA socket, you will just measure an open circuit. If you open the case you can then put one test lead just after the first capacitor (you'll need to trace it, ignore the disc caps, it will be film or electro), and one lead on the RCA socket outer terminal.
This will give a close practical approximation in most cases. It is better to get the spec from the manufacturer.
Failing that, just use a small value cap, and measure or listen for frequency response roll off. No harm really.
Hi,
I don't know the specifics of your amp, but I bet it has one of those AC mains filters integrated in the power receptacle, or a power filter lurking somewhere else inside. These usually consist of a bifilarly wound choke, and so called X and Y capacitors. The X capacitors crossover from one power wire to the other, in order to short out HF differential mode interference, and the Y caps (there are two of them) go from each power wire to ground. They should short out common-mode interference. If your safety earth is not connected, then those Y capacitors form a capacitive voltage divider which actually put 1/2 of your mains voltage on the circuit ground! Fortunately, thes caps are safety rated and of rather small value, so the current conducted by them is only enough to give you a nip if you happen to touch both circuit ground and "true" earth.
If you measure the voltage between your amp and your window sill, then there is a severe impedance in series with 1/2 the mains voltage, being the reactance of the Y caps, and the impedance from your window sill to Earth. It may be grounded by contact with rebar, but much more likely, it has a high impedance to Earth through moist stone/concrete/wood or whatever your house is made of. When you load it with an analog multimeter, which is not very high in impedance, the voltage you measure may only be a small fraction of the voltage that is really present.
So if one of those mains filters is present, you should always connect safety earth, or you'll face a serious (and practically uncurable) hum problem with possibly disastrous results. You may only disconnect safety earth if your power transformer, and all your mains wiring inside the amp confrom to standards of double insulation. This is a rarity in tube gear.
Regards,
Jurgen
I don't know the specifics of your amp, but I bet it has one of those AC mains filters integrated in the power receptacle, or a power filter lurking somewhere else inside. These usually consist of a bifilarly wound choke, and so called X and Y capacitors. The X capacitors crossover from one power wire to the other, in order to short out HF differential mode interference, and the Y caps (there are two of them) go from each power wire to ground. They should short out common-mode interference. If your safety earth is not connected, then those Y capacitors form a capacitive voltage divider which actually put 1/2 of your mains voltage on the circuit ground! Fortunately, thes caps are safety rated and of rather small value, so the current conducted by them is only enough to give you a nip if you happen to touch both circuit ground and "true" earth.
If you measure the voltage between your amp and your window sill, then there is a severe impedance in series with 1/2 the mains voltage, being the reactance of the Y caps, and the impedance from your window sill to Earth. It may be grounded by contact with rebar, but much more likely, it has a high impedance to Earth through moist stone/concrete/wood or whatever your house is made of. When you load it with an analog multimeter, which is not very high in impedance, the voltage you measure may only be a small fraction of the voltage that is really present.
So if one of those mains filters is present, you should always connect safety earth, or you'll face a serious (and practically uncurable) hum problem with possibly disastrous results. You may only disconnect safety earth if your power transformer, and all your mains wiring inside the amp confrom to standards of double insulation. This is a rarity in tube gear.
Regards,
Jurgen
Hi senderj,
The good news is that any small signal relay will work fine. Use a standard 2C contact form signal switching relay. Forget about the capacitor problem and just time it. This will reduce the time somewhat as it shorts the output to ground. Same as your amplifier having maybe a 1 ohm input impedance.
Sound good?
-Chris
The good news is that any small signal relay will work fine. Use a standard 2C contact form signal switching relay. Forget about the capacitor problem and just time it. This will reduce the time somewhat as it shorts the output to ground. Same as your amplifier having maybe a 1 ohm input impedance.
Sound good?
-Chris
GROUND VS EARTH
Many amplifier circuits have the circuit ground at a different potential than the chassis ground. In most cases there will be either a resistor or capacitor or usually both, between the circuit ground and chassis ground. If you connect the circuit ground to the chassis ground you will have major problems. Do Not do it!
If you do not understand the design well enough to see the reasons for this, you are certainly not competent to understand the consequences of modifying it, and they are serious and even possibly catastrophic (such as the output stage overheating and frying components, for example). Again....DO NOT DO IT!!!!
Many amplifier circuits have the circuit ground at a different potential than the chassis ground. In most cases there will be either a resistor or capacitor or usually both, between the circuit ground and chassis ground. If you connect the circuit ground to the chassis ground you will have major problems. Do Not do it!
If you do not understand the design well enough to see the reasons for this, you are certainly not competent to understand the consequences of modifying it, and they are serious and even possibly catastrophic (such as the output stage overheating and frying components, for example). Again....DO NOT DO IT!!!!
If I have 3 separate ground in my amp:
- signal grnd
- small PS (for input stage + vas) gnd
- big PS (output stage and speaker) gnd
1) Is it possible to connect them only through a resistor ?
(left one is a dac, right one is the amp)
2) Where (to wich ps-gnd) should I connect the gnd points in the amp ?
(input resistor, feedback resistor, current-source, dc servo, etc...)
3) Attachment: its a correct gnd-ing between 2 devices ?
Thx !
- signal grnd
- small PS (for input stage + vas) gnd
- big PS (output stage and speaker) gnd
1) Is it possible to connect them only through a resistor ?
(left one is a dac, right one is the amp)
2) Where (to wich ps-gnd) should I connect the gnd points in the amp ?
(input resistor, feedback resistor, current-source, dc servo, etc...)
3) Attachment: its a correct gnd-ing between 2 devices ?
Thx !
I have recently built this DAC : http://i12.photobucket.com/albums/a250/ThomasC2/DAC1.jpg
As can be seen , I used 6 small cheap power transformers. It kept on blowing fuse when I first turned it on (while chassis was earthed). When I disconnected the chassis from earth, it stopped blowing fuse. Eventually, I found over 80V a/c on the chassis (when compared with earth)!!! I tried reversing the connections of the primary sections of the transformers and I found such voltage on the chassis varied by about 10V. I then insulated all the transformers from the chassis by placing a piece of plastic between the transformers and the chassis and such voltage was lowered to below 30V.
I still could not understand where such 30V comes from. Later I will replace the 6 individual transfrormers with a more decent one and see.
As can be seen , I used 6 small cheap power transformers. It kept on blowing fuse when I first turned it on (while chassis was earthed). When I disconnected the chassis from earth, it stopped blowing fuse. Eventually, I found over 80V a/c on the chassis (when compared with earth)!!! I tried reversing the connections of the primary sections of the transformers and I found such voltage on the chassis varied by about 10V. I then insulated all the transformers from the chassis by placing a piece of plastic between the transformers and the chassis and such voltage was lowered to below 30V.
I still could not understand where such 30V comes from. Later I will replace the 6 individual transfrormers with a more decent one and see.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.