Hiya folks,
I'm starting on my first DIY, a PP 7591 based stereo amp using 6AN8s for gain/phase inverter(standard cathodyne), 2 5AR4s for power and a pair of 0A3s for screen regulation.
I got some grounding questions cleared up with a search, but have a few more:
I'm using an earth ground with mains grounded to chassis as soon as it enters the amp. From there it's "multiple star grounds" floating on terminal strips with pwr xfrmr CT and first filter caps as well as filament CT forming the circuit star ground start.
Next up is filtering for left and right channel-a separate star for each channel, each star running a ground back to circuit ground start. Picture a "V" with the base of the "V" as the circuit star ground start and successive left and right channel stars as the left and right side of the "V." Hope that made sense!
Next, left and right ground stars for the outputs, running a ground line back to the previous ground star for left and right channel-the left/right filter section ground stars.
Next up, left and right ground stars for the phase/gain sections-each running a ground back to the previous ground stars-the outputs.
RCA inputs are insulated and will ground to the 500k stereo attenuator, which in turn will run a ground to the phase/gain ground stars.
Is this bus now in fact acting as a "funnel" for ground current and voltage, going from low-voltage low-current to progessively higher voltage and higher current as each ground star makes its way back to the circuit star ground start?
As for that circuit star ground start, the plan was to run it to from the terminal strip on which it starts directly to mains/earth/chassis ground shared by the mains-but I've read that mains/earth/chassis ground should have its' own separate ground.
Is this true, and if so-why? From my reading this would be the definition of a ground loop-a circuit that references more than one ground at different potentials, right?
Any info would be most appreciated.
Best,
PP7591
I'm starting on my first DIY, a PP 7591 based stereo amp using 6AN8s for gain/phase inverter(standard cathodyne), 2 5AR4s for power and a pair of 0A3s for screen regulation.
I got some grounding questions cleared up with a search, but have a few more:
I'm using an earth ground with mains grounded to chassis as soon as it enters the amp. From there it's "multiple star grounds" floating on terminal strips with pwr xfrmr CT and first filter caps as well as filament CT forming the circuit star ground start.
Next up is filtering for left and right channel-a separate star for each channel, each star running a ground back to circuit ground start. Picture a "V" with the base of the "V" as the circuit star ground start and successive left and right channel stars as the left and right side of the "V." Hope that made sense!
Next, left and right ground stars for the outputs, running a ground line back to the previous ground star for left and right channel-the left/right filter section ground stars.
Next up, left and right ground stars for the phase/gain sections-each running a ground back to the previous ground stars-the outputs.
RCA inputs are insulated and will ground to the 500k stereo attenuator, which in turn will run a ground to the phase/gain ground stars.
Is this bus now in fact acting as a "funnel" for ground current and voltage, going from low-voltage low-current to progessively higher voltage and higher current as each ground star makes its way back to the circuit star ground start?
As for that circuit star ground start, the plan was to run it to from the terminal strip on which it starts directly to mains/earth/chassis ground shared by the mains-but I've read that mains/earth/chassis ground should have its' own separate ground.
Is this true, and if so-why? From my reading this would be the definition of a ground loop-a circuit that references more than one ground at different potentials, right?
Any info would be most appreciated.
Best,
PP7591
Grounded for life
Because the mains ground is needed for safety this ends up complicating the question. (Although you would hardly be the first to use a two wire power cord.)
With a transformer powered design there is no particular need for the mains ground and the circuit ground to have any current flow between them at all and in fact there should be as little as possible. The only small exception is transformer interwinding capacitance (near zero for toroids) causing some small AC common ripple current
The ground currents on the amplifier side of the transformer flow circularly between the speaker return path and power supply and between the input connector (RCA or XLR?) and the signal return reference (input stage). These currents should not be allowed to interact and a star system is good for this.
The previous box in the chain has probably connected their RCA return to their chassis taking part of the decision out of your hands.
But if you visualize two seperate sets of circular ground currents one on the primary and one on the secondary side of the transformer it may help you resolve your opening question
Because the mains ground is needed for safety this ends up complicating the question. (Although you would hardly be the first to use a two wire power cord.)
With a transformer powered design there is no particular need for the mains ground and the circuit ground to have any current flow between them at all and in fact there should be as little as possible. The only small exception is transformer interwinding capacitance (near zero for toroids) causing some small AC common ripple current
The ground currents on the amplifier side of the transformer flow circularly between the speaker return path and power supply and between the input connector (RCA or XLR?) and the signal return reference (input stage). These currents should not be allowed to interact and a star system is good for this.
The previous box in the chain has probably connected their RCA return to their chassis taking part of the decision out of your hands.
But if you visualize two seperate sets of circular ground currents one on the primary and one on the secondary side of the transformer it may help you resolve your opening question
Disclaimer- I am not an expert having only built a couple of tube amps.
Having said that your grounding system so far sounds like what I used with my monoblocks. The power side has a common ground point and the signal side has its own ground on a terminal strip.
I then ran two wires, one from each of those two ground points to the safety ground right where the AC connects to the chassis. The result is two dead quiet amps, no hum from them at all.
I agonized over how to connect the grounds and finally decided to just make a decision and go with it. After all it is DIY and I could change it later if it didn't work. In my case it worked perfectly.
Having said that your grounding system so far sounds like what I used with my monoblocks. The power side has a common ground point and the signal side has its own ground on a terminal strip.
I then ran two wires, one from each of those two ground points to the safety ground right where the AC connects to the chassis. The result is two dead quiet amps, no hum from them at all.
I agonized over how to connect the grounds and finally decided to just make a decision and go with it. After all it is DIY and I could change it later if it didn't work. In my case it worked perfectly.
Hey sherman and herman-thank you both for your help; much appreciated!
Nice to know I'm at the very least on the right track. When researching ground schemes about the only constant I found was "use black insulated wire for ground!" Everything else seemed up for grabs.
I know by the very nature of CT transformers in a PP circuit I have the advantage of cancelling some hum, and in fact should have no current going to ground at all on the CTs, but it would be nice to not have to cancel any at all if possible.
Sherman,
Did you build single-ended monos or PP? Either way, what'd you use for a tube line-up?
Also, I went back and re-read the info on star-grounding, and after some reflection I don't think they were implying the mains ground had to be separate, just that it had to have it's own ground point to the chassis, period. It would indeed make sense to ground the circuit side to the mains/earth/chassis ground just as you did-no potential for a ground loop; at least within the amp circuit-if both sides (AC and circuit) both reference the same point for ground.
A big reason for my DIY is to build something unique, build something much better than I can buy for any reasonable amount of money and perhaps the most important reason-to gain a deeper understanding of what's going on and why.
Guess that's what drives most of us to pursue this hobby!
Best,
pp7591
Nice to know I'm at the very least on the right track. When researching ground schemes about the only constant I found was "use black insulated wire for ground!" Everything else seemed up for grabs.
I know by the very nature of CT transformers in a PP circuit I have the advantage of cancelling some hum, and in fact should have no current going to ground at all on the CTs, but it would be nice to not have to cancel any at all if possible.
Sherman,
Did you build single-ended monos or PP? Either way, what'd you use for a tube line-up?
Also, I went back and re-read the info on star-grounding, and after some reflection I don't think they were implying the mains ground had to be separate, just that it had to have it's own ground point to the chassis, period. It would indeed make sense to ground the circuit side to the mains/earth/chassis ground just as you did-no potential for a ground loop; at least within the amp circuit-if both sides (AC and circuit) both reference the same point for ground.
A big reason for my DIY is to build something unique, build something much better than I can buy for any reasonable amount of money and perhaps the most important reason-to gain a deeper understanding of what's going on and why.
Guess that's what drives most of us to pursue this hobby!
Best,
pp7591
Tube PP Topologies
Hi PushPull7591;
(Are there really 7591 previous users of PushPull?)
I have no idea how far along you are in your design, I ran into an interesting paper on the internet the other day about tube topologies. Well written by Menno van der Veen with the right amount of usefull information. There's no date so this could easily be old news but I thought some very good ideas were presented. Perhaps you'd like to review it.
http://www.plitron.com/PDF/PB/Article/Atcl_4.pdf
Hi PushPull7591;
(Are there really 7591 previous users of PushPull?)
I have no idea how far along you are in your design, I ran into an interesting paper on the internet the other day about tube topologies. Well written by Menno van der Veen with the right amount of usefull information. There's no date so this could easily be old news but I thought some very good ideas were presented. Perhaps you'd like to review it.
http://www.plitron.com/PDF/PB/Article/Atcl_4.pdf
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.