Seconded, his data is outstanding, and his books are great, not only does he point out that star grounding is very effective, but other types of very successful grounding schemes.
I agree on Merlin's grounding. I had to go back to his article several times(I'm a little slow) but once I got it, it made perfect sense! My interpretation is that it's all about controlling where currents travel and thinking about each node in the power supply as a separate power supply. Now I have to go back and rebuild my first builds!
I have read multiple times on this forum and other forums that the preamp should be grounded at the input jack. Does that mean that the preamp ground is through the mechanical connection of the input jack? To make a good ground, you would have to tighten the jack pretty tight to get a good enough connection to the chassis. If you use a Plexi face plate, it seems like that might crack the faceplate. Also, the cliff jacks I have been using are isolated, so there would be no connection to the chassis if I use those jacks.
What I have been doing is to drill a hole for the preamp ground buss as close to the input jack as possible and then connect the ground of the input jack to that ground buss with 18 gauge solid copper wire.
Is this OK? I can't see any other way if the input jack is isolated. Any other suggestions?
In most cases the cliff jacks are insulated and they won't make a connection. Also the jacks themselves, even if they do, often wiggle loose over time, and build up corrosion that can make that ground weak at best if not have it removed totally. The best method is to create a drill hole with a solder tab type connection on the chassis right next to the input jacks as close as physically possible and make that the ground connection. the input jacks themselves can still be a 'redundant' ground at the same spot, but if they wiggle loose, it won't harm things, and having two grounds within less than an inch is pretty close to the same in my experience.
mwelch55 wrote: ↑Tue Jul 17, 2018 5:53 pm
I have read multiple times on this forum and other forums that the preamp should be grounded at the input jack.
I have posted about this previously. I am the outlier here. I never ground at the input jack, yet, I get good results. I was taught by someone (online) who I respected, to run a buss from the input to the PI, attaching ground wires clustered by gain stage. Then from the PI end, ground at a single chassis lug that also has the high potential grounds (power cathodes, PT CT, main filter, speaker, etc.) It works for me. I'm not looking for debate on it. It's a simple point, there is more than one way to get a good result. Please do what works for you.
mwelch55 wrote: ↑Tue Jul 17, 2018 5:53 pm
I have read multiple times on this forum and other forums that the preamp should be grounded at the input jack.
I have posted about this previously. I am the outlier here. I never ground at the input jack, yet, I get good results. I was taught by someone (online) who I respected, to run a buss from the input to the PI, attaching ground wires clustered by gain stage. Then from the PI end, ground at a single chassis lug that also has the high potential grounds (power cathodes, PT CT, main filter, speaker, etc.) It works for me. I'm not looking for debate on it. It's a simple point, there is more than one way to get a good result. Please do what works for you.
Phil S, That's the way I have done it too with excellent results, and an engineer I respect very much with 50+ years in the audio industry told me to do it that way too. I think there are likely many ways to get good results.
Hi all, FWIW I updated to fill in the missing diagrams (tried to link but for some reason that didn't work so had to upload them). Been meaning to do this--sorry it took so long! Hopefully the article makes more sense. Sadly Mr. Kimura passed away sometime last year from ALS (otherwise known as Motor Neuron disease or "Lou Gehrig's disease"(RIP). (Note that the original was altered since I did my translation so the final version on his site is a bit different from the one here.)
mwelch55 wrote: ↑Tue Jul 17, 2018 5:53 pm
I have read multiple times on this forum and other forums that the preamp should be grounded at the input jack.
I have posted about this previously. I am the outlier here. I never ground at the input jack, yet, I get good results. I was taught by someone (online) who I respected, to run a buss from the input to the PI, attaching ground wires clustered by gain stage. Then from the PI end, ground at a single chassis lug that also has the high potential grounds (power cathodes, PT CT, main filter, speaker, etc.) It works for me. I'm not looking for debate on it. It's a simple point, there is more than one way to get a good result. Please do what works for you.
Phil S, That's the way I have done it too with excellent results, and an engineer I respect very much with 50+ years in the audio industry told me to do it that way too. I think there are likely many ways to get good results.
If you are using isolated input jacks, run a 0.01uF cap from the ground/sleeve connection directly to the chassis with as short a leads as possible.
Elsewhere he explains that he learnt this through bitter experience, in which a customer in an area with high EMI experienced problems that were solved by the cap.
The wire inductance between the input and the chassis screen is the cause.
Phil_S wrote: ↑Mon Sep 19, 2022 8:56 pm
I have posted about this previously. I am the outlier here. I never ground at the input jack, yet, I get good results. I was taught by someone (online) who I respected, to run a buss from the input to the PI, attaching ground wires clustered by gain stage. Then from the PI end, ground at a single chassis lug that also has the high potential grounds (power cathodes, PT CT, main filter, speaker, etc.) It works for me. I'm not looking for debate on it. It's a simple point, there is more than one way to get a good result. Please do what works for you.
Phil S, That's the way I have done it too with excellent results, and an engineer I respect very much with 50+ years in the audio industry told me to do it that way too. I think there are likely many ways to get good results.
If you are using isolated input jacks, run a 0.01uF cap from the ground/sleeve connection directly to the chassis with as short a leads as possible.
Elsewhere he explains that he learnt this through bitter experience, in which a customer in an area with high EMI experienced problems that were solved by the cap.
The wire inductance between the input and the chassis screen is the cause.
Yes Aiken is usually right and has a lot of great advice. That is likely the best way to do it with most applications as anything coming into the amp will short out right away instead of going through and getting amplified. Having said that though, this amp I was working on that my engineer friend advised me to ground in the fashion you described instead of at the input jacks was quieter and more stable when not grounded at the input jack. I tried both approaches. This amp is unique in that it is a former Conn organ power amp chassis that I modified. It has three separate power amps, phase inverters, and output transformers in it, all sharing a common power transformer. I made a separate chassis that has three separate preamps in it. Each preamp grounds to it's own buss in order of the stages, with local stars that each cap supplies, and those busses continue via an 11 pin socket that passes into the power amp chassis, where there is a separate ground buss for each amp. They all sum at the same chassis ground point. Every stage in the three amps are decoupled with their own filter caps. It all fits into one head, with 15 tubes (one more than a vintage SVT!). I have to use it with a Weber Four Head (no longer sold) unless I want to carry around 3 cabinets all the time.
I had some trouble with it not surprisingly, with preamps interacting overly and instability. It is an overly complicated project for sure, but after 3 revisions to the preamp chassis layout and design I got it stable and it sounds really good and works well. Conn used Foster transformers (Cincinnati) and way oversized them, and it sounds great as a result. I talked with Andy Marshall from THD about it when I was first working on it and he told me Foster transformers were some of the best that were on tube audio gear. He also thought I was bonkers and a glutton for punishment. Best of all I learned a lot with the project.
dragonbat13 wrote: ↑Sat Dec 24, 2022 10:54 pm
Phil, have you built any high gain amps with this ground at the PT end?
While the PT ground works, the idea is that it lowers the noise floor at the input side of the chassis.
Are your amps noise free?
Just seeing this now...if that matters. I'm not a fan of high gain. Perhaps the highest gain amp I'm built is an Express variant of my own thinking, so not really high gain. The noise level in all of my amps is quite descent. Not all are quiet as a mouse, but all quite satisfactory in terms of noise. Your comment about noise floor is appreciated in that now I have the reason why people ground at the input.