Ground Loops

Updated with new material 7th January 2019

This set of c. 70 slides is the culmination of my experience over a period of about 25 years building numerous power amplifiers and preamplifiers. I first started out in audio around 1975 or 76 a teenager. Some of my creations were reasonably quiet – through pure luck – and others hummed and hissed horribly. Later, my skills improved dramatically, and especially so after reading one of Henry Ott’s books back in about 1988/89 whilst developing a very high resolution Digital Panel Indicator for industrial applications at the company I worked for. I then left DIY audio for about 15 years (career, family etc), returning to the subject again about 15 years ago, having forgotten a lot of my practical skills. The path from electromagnetic theory expounded on numerous websites, application notes and posts on various web forums to building quiet amplifiers every time is not easy and requires a bit of practice. The underlying theory can be extremely complex (think Maxwell’s equations), however, with some effort and focus you can quickly master the basics.  This set of slides focuses on unbalanced interconnects (aka ‘single-ended’) that use the standard RCA phono connectors, since this is where problems mostly arise.

When it comes to humming and hissing amplifiers, good practical advice is scarce, though often misguided opinions on the subject seem easy to find.

This presentation (which will remain a work in progress and be expanded from time to time) is designed to get audio constructors up and running quickly both in the construction/planning phase, but also debugging. It will hopefully also serve as a useful reference for anyone wanting to know a bit more about EMC as applied to amplifiers.  One important thing about EMC: you will never stop learning, and finding new problems to solve.

Ground Loops

Here is DIYaudio member Ilimzn’s excellent posts on the subject that I gathered into a single document

ilimzn’s Excellent Posts on Ground Loops

Here is some additional material

Amplifier PCB Design Guidlines for Minimizing Hum

Some practical guidance offered to a builder on diyAudio

More Notes On Amplifier Hum Problems

For some practical examples of low noise amplifiers using these techniques,  see the nx-Amplifier and sx-Amplifier on www.hifisonix.com

Here are some commercial products that use the techniques described in the presentation: www.ovationhifidelity.com

The picture below shows the internal wiring of one of the two nx-Amplifiers I built with zero noise or hum problems. The power wiring is tightly bundled, and small signal wiring is kept well away from the transformer and other power wiring. On the PSU PCB, strict attention to the capacitor ‘T’ connection and ‘star’ ground return result in an exceptionally quiet amplifier.


IMGP8749

Finally here is a YouTube video of HOW NOT TO DEAL WITH GROUND LOOPS. What is really disconcerting about this is that if you type ‘Ground Loops’ into Google, this will probably be the first reference that comes up on the list.

All the basic safety rules about earthing [grounding] by using a ‘ground lifter’ are completely broken and the noise problem has actually not been solved – they have gone around the problem and made the product completely unsafe in the process – and especially so since this is on a tube amplifier.

The question to ask when dealing with a potential safety fault is ‘What would happen if the live wire  came loose and touched on the input jack, or any other metalwork on the product without the safety earth[ground] connected?’ If the answer is ‘it would be at the live potential’ don’t do it!

Never, ever, use a ground lifter in the manner shown to get rid of hum – its plain dangerous and illegal. Period.

Comments

53 responses to “Ground Loops”

  1. Lai Chang Lin says:

    Thanks for sharing the information. It is very useful to me because I build apmplifiers as a hobby and the the ground loop noise has always troubled me.

    Do you live in Taiwan? It strikes me as pretty odd that an American doing amplifier DIY in Taiwan. Your builds look beautiful, much better than mine.

    • Bonsai says:

      Hello LC – yes I live in Taiwan (3 years now) . I have been many times to Delta on business.

      Glad you like the website. Any questions, just let me know.

  2. Paul says:

    Hi Bonsai, this is Shredhead from diy. This site is super cool. You must have some photography chops too because the pictures here are awesome. I was wondering if you might someday get to a “How to Wire up a Preamp” paper. I am trying to wire one up using a star ground method because I am not using a PCB for the preamp board with a ground plane in it. I am getting all kinds of different advice from DIY on how to run the ground and it kind of makes me think that star ground isn’t the way to go for line levels. Is this right?

  3. Paul says:

    http://www.diyaudio.com/forums/analog-line-level/262728-one-these-wiring-schemes-will-have-lower-noise.html

    I have this thread about it right now on DIY. I have changed the layout today and that is on the 3rd page. Sorry about all of the safety ground arguing on there, I am just frustrated that no one seems to understand the best way to wire things. I will follow your advice and a couple of other people to the letter but it’s hard to know who else to trust.

  4. LC says:

    Did you gcome to Delta headquarter in Taipei or other locations? I am located in Taipei. How do like living in Taiwan?

    Do you mind telling me where you get your PCB made?

    • Bonsai says:

      Sorry for late reply LC.

      Yes – I enjoy it in Taiwan!

      I get my PCB’s made by PCBcart who are based in China – very good service and price is resonable. I never had any quality or delivery problems from them and you can pay with Paypal. They take Gerbers.

  5. LC says:

    Hi Bonsai,

    Thanks for the information on PCB maker.

    I have a question about hum in power amplifier. In the amplifier I built I used one transformer but two capacitor banks, one for each channel. I always get hum on one of the channels. I connected the gounding as described in your paper, but still get hum on one channel. Any suggestions? Thanks!

    Best Regards,

    LC

    • Bonsai says:

      LC, I have had the same problem.

      You really have to be very careful with potential ground loop problems between the channels and common impedance coupling. Further, because you are feeding two separate capacitor banks, if youndo not take care, the EM radiation from the capacitor charging currents is is also greater.

      A much more reliable method if your transformer has split secondary windings is to rectify and smooth each secontary and them combine them to form a +- DC supply, remembering of course to ‘T’ off from where you join them to make the star ground. I would suggest that you do it this way.

  6. WillyBoy says:

    Hi Andrew,

    I read this particular post with great interest, as I am finishing my DIY amplifier and experiencing some issues with hum. With input connected but volume is on zero, there is hum and audible radio broadcast interference, which from your post I take as RFI. I also tried coupling R and L input with a cable and the hum is still there, not getting worse but it’s there. Could be a case of cross channel ground loop?

    Remedy:
    1. At the RCA input, put 1500R in series and 220pF parallel to ground before the cable that connects to the amp board. This is low pass filter that filters out RF above 480kHz.
    2. At the same location, I will put 1.8nF in series from the common ground (I joined the sleeve tab for R and L RCA together) to the metal enclosure via a small bolt. Question: Can I connect this will somewhat longer cable and connect it at the star ground point? Around 30cm or so.
    3. For the hum breaking resistor, where do you connect it on the amp board, and connect to main star ground point? Sorry I am not 100% sure when it comes to reading schematics :).

    I am keen to learn and willing to try the suggested remedies, hopefully I can get it working smoothly without any hum or issues. Thank you for sharing your experience and expertise with us here and on the forums.

    • Bonsai says:

      Willyboy,
      Follow these steps to try to locate the cause of the problem:-

      1. Short the input of the amplifier right where the signal enters the PCB. If it still hums, you probably have a common impedance problem either with the on-board decoupling and signal routing, or with the main powerr supply caps. With the input shorted (or open and without any input cable connected) an amplifier should be completely quiet.

      2. You may also have noise coupling into the amplifier circuit through stray magnetic fields from the transformer. You can check if this is a problem by moving the transformer (or the amplifier module) away. If the hum decreases, its likely you have this problem – if it hardly changes – more likely #1

      3. Make sure your input connector(s) ground is not making contact with the chassis – if it is, you will have hum.

      You need to get this sorted out first, before you look into any other potential mechanisms (cross channel ground loop, RFI)

  7. WillyBoy says:

    Hi Andrew,

    Thank you for getting back to me. I tried what you advised, and found the following:

    1. No input cable connected, amp is silent. With input cable connected but a jumper from L to Ground (I only tested one channel) it is also silent.

    2. I tried moving the tone control, the power amp, the potentiometer, hardly changes anything.

    3. As far as I know they’re well isolated.

    I’ve also installed 1k8 Ohm in series and 220pF parallel to ground on the input directly to the RCA sockets, this hardly changes anything.

    Can you help me more with this information? Thank you also for using simple language and practical approach.

    • Bonsai says:

      ok – that’s a good start – looks like you don’t have a common impedance problem good.

      When you say you’ve connected a 1.8k and 220pF in parallel to ground, I assume that is on the amplifier module and not to the amplifier chassis – can you confirm that my understanding is correct?

      If you connect a cable from the left channel to the right channel what happens?

  8. WillyBoy says:

    Hi Andrew,

    I connect the 1k8 and 220pF to form an RC low pass filter at the input. I install them on RCA socket R and L and the other leg of the two caps are joined together on the (also joined between R and L) sleeve tab, meets the signal ground cable. This is not connected to safety ground.

    Right now when I plug in an RCA jack on the L, there is a radio broadcast noise but barely intelligible. When I plug the other end of that cable to the R socket, the noise seems to get louder, dare I say twice as loud and makes me think more than one station is being received. This is with volume pot all the way down. As I open it the noise goes away.

    Hope that makes sense. If you want to see some pictures I can send them to you. Thanks again for handholding through this process.

    • Bonsai says:

      Ok. Looks like an RFI issue definitely. Can you put a 2-5 nF cap from each input socket ground to the chassis right where the connector is mounted. At RF this makes the screen and the amplifier one enclosure. You will need to use a ceramic cap for this and keep the leads really short. Leave your input filter as you have it.

  9. WillyBoy says:

    That did it- really helped. I used 1.8nF for it and the radio noise for sure went down. With input connected there is still some hiss with the volume all the way down. Is there anything I can try or should I just close it up and be happy? Thanks so much.

    • Bonsai says:

      The other thing you can try is to do the same on the speaker connections. RF can also come in via the + terminal.

  10. WillyBoy says:

    I also noticed when input cable is connected and volume all the way down, when I touch the heatsink the hum is reduced. Is there something I can do here, Andrew? Thanks again..

    • Bonsai says:

      Hi,

      firstly, make sure your amplifier is earthed, and all the metal components (including the heat sinks) are also well earthed.

      There is a good chance that your amplifier may be oscillating if you are getting this problem. What happens in this case is when you touch the amplifier, you change the capacitance to ground (whether its earthed or not since at HF the wiring inductance will play a part) and this alters the oscillation frequency or the oscillation amplitude. The transformer magnetic field modulates the HF oscillation which then gets rectified by the semiconductor junctions in the amplifier and the result is hum or buzz (demodulated HF oscillation). This also may well be the cause of your earlier reported RFI problem.

      You need to put a scope on the output and make sure its stable. If you don’t have a scope, make sure you have the correct compensation components in place. Does your amplifier have an output coil and a Zobel network in place (it should!) and check the values are correct – these are all factors that can lead to instability.

  11. Michael Cookson says:

    I have a question on a hum I have in my diy amplifier. It uses a large toroidal transformer, two yuan-jing 2×150 amp boards. I am using it to bi amp my 3way passive studio monitors. The monitors use a 7″ dual concentric speaker from Seas that has its own passive xover, and an 8″ Seas woofer for bass that I use a minidsp to crossover the two woofers. 1 amp powers the mid/highs, 1 amp for the lows. Now to the hum. When I have all 4 inputs plugged in I get a hum, when I have either the mid/highs or lows plugged in by themselves there is zero hum. Any thoughts on the cause? Thank you

    • Bonsai says:

      Michael,
      not easy for me to answer without seeing the layout etc. Do all the inputs come out of a single source box, or do you have two separate sources? If the amplifiers are silent without any input whilst connected to the speakers, I’d say you seem ok insofar as the internal wiring goes – i.e. no common impedance problems etc which would manifest even without an input. If I understand you correctly, then its only when you connect the second set of inputs that you get the problem. This looks like you have either a cross channel ground loop or you are creating a new ground path when you connect the second set of inputs through the source equipment, thereby creating a large loop area.

      If you have the inputs plugged in, the amplifier powered up but the SOURCES powered down, does it still hum?

      Make sure your input connectors do not make contact with the metal chassis where they enter the housing (although, given that the amp is quiet without inputs, I don’t believe this should be a problem – but just check in any event.

      As a first step, I’d look at you amp and determine if the modules have a suitable hum breaking resistor installed – if not, you need to fit them (one in each channel).

  12. Hi Bonsai,

    Just would like to say thanks for the information. I’ve been rebuilding my 5 channel pre/power amplifier and had problems with hum. The layout isn’t ideal but I actually found that separating the signal / power grounds on the amplifier PCBs I had (LM3886) and installing a hum breaking resistor at each amp made the world of difference!

    I had to cut traces to do it – but very pleased with the result. I also send all signals ground direct to each amp now instead of the PSU star/T. Only the power ground, speaker return and protection ground are at the PSU T now. Now I have to put my ear to speaker to hear any noise at all – and that is only a very slight hiss on the tweeter. Very happy.

    • Bonsai says:

      That’s fantastic Daniel – Its great when you can get an amp really quiet.

      I’ll be making a few updates to the presentation in the next few weeks – there’s a few things I need to make clearer and that can improve the noise further.

      Anyway – great that you’ve had a good result!

  13. ahmed says:

    Thanks, I made 2 hobby 5.1 amps and both sufered from cross channel ground loop, DIY community gives little to no attention to this. solved my hum issue

  14. Mikkel Bifrost says:

    Hello.
    I salute you for your work and the sharing of your knowledge. With great sound comes great responsibility!
    I am still new to this, the whole construction aspect, but i cant stop until i understand the different aspects of the whole chain. from source to speakers.

    I was wondering if you have any knowledge of, or experience with, what is called a r-core transformer? It should supposedly be the best choice for powering an audio amp, since the magnetic field is completely contained within the transformer, so none of the wires or circuitry will be adversely affected.

    And can you maybe answer, if you know, what the deal is with using balanced and unbalanced components in the same chain? Will that not be defeating the whole purpose of making a chain balanced?

    Again thank you for sharing.
    Thank you for being, thank you for you.
    May light be ever illuminating your path.

    With loving regard,
    Mikkel Bifrost

    • Bonsai says:

      Hello Mikkel,

      Thank you for your kind words. I am not an expert in transformers, but from what I do know:-

      Toroids offer the highest volume vs power efficiency and the lowest radiated magnetic field. They also have very low core losses. However, since the secondary is usually wound over the primary, the interwinding capacitance is very high, and this allows noise (common mode and series mode) to capacitively couple across the windings. The power bandwidth of a typical 500-1kVA toroid is about 60 kHz in series mode, and the capacitive coupling up to 10’s of MHz (common mode). To get around the capacitive coupling problem, you can specify an interwinding screen, which can reduce the interwinding capacitance by a factor of 10x. You can also specify a GOSS band (aka flux band or ‘belly’ band) that can further reduce any radiated magnetic field to very low levels. For general purpose audio, including small signal work, I would always recommend a toroid but with an interwinding screen and a GOSS band. Note, that if you want low mechanical noise, you should specify a ‘stabilized’ core. With this type of core, the toroid laminations are vacuum impregnated prior to winding with a special varnish to acoustically dampen the effects of magneto-restriction. This is when the physical dimensions of the core changes due to the magnetic force exerted on them during excitation.

      R core transformers are not as volumetrically efficient as toroids, but they have very low interwinding capacitance and are also suitable for high galvanic isolation applications.

      EI cores are the oldest style transformer and still in use in high volumes. However, they suffer from serious magnetic field radiation problems, so for low noise applications like audio, you have to be careful about this aspect. They can be fitted with a GOSS band that will reduce this problem, or fitted in a metal can (often you see this in valve equipment), but they will never be better than toroids in this regard.

      Balanced vs unbalanced – idealy you should go balanced if you can afford it, or you can get equipment (source and power amplifiers) that offer this feature. However, most domestic and DIY systems are still unbalanced, and this is why I focused on it in my ‘Ground Loops’ article. If you do a good job on an unbalanced system, given the short cable runs in a typical domestic systems, you can still get very good results.

  15. Veysel arslan says:

    Nice documents ty Bonsai

  16. […] på 100Hz. I tillegg var det en del annen støy. Etter å ha lest litt i pdf fila som ligger på hifisonics så flyttet jeg litt på noen kabler, flettet sammen power kablene mellom CRC poweret og […]

  17. […] and noise in amplifiers and their relationship to grounding, with considerable practical advice, this post and associated material from the HiFiSonix website is […]

  18. John says:

    I’m really enjoying your articles on ground loops. I have an unusual situation. I’m fabricating four amplifier chassis/components to create a stereo amplifier. I’m doing this because the power supply components are heavy, and it’s easier to handle the individual chassis when I separate the mosfet power supplies (low voltage/high current) from the vacuum tube/filament power supplies (higher voltages and filament voltages).

    My question is this: how do you properly ground four chassis to earth? I’m getting a slight amount of 60 hz hum, but I want zero hum at the speaker. You mention using only one ground to earth, but I have four chassis (only three are connected to 120 volts AC). The amp is receiving the DC from the three power supplies.

    I’ll be adding the 15 ohm resistors to the source ground, as you mentioned. I’ll also redo the wiring (twist + and ground wires together).

  19. Cuong Nguyen says:

    Hi Andrew,
    I read your article many time, so i have a question below.
    With stereo amp, speaker return is routed to amplifier PCB 0V. So, how about dual mono amps/monoblock amps? Speaker return should be placed at amplifier PCB 0V or PSU 0V?
    Can you explain for me? Thanks!

    • Bonsai says:

      Hello Cuong,

      Dual MONO is difficult problem with unbalanced inputs! I have not built an amplifier like this, but my suggestion is:-

      1. Build each amplifier following the normal rules. Keep loop areas small, tightly twist all wiring, return speaker 0V to amp board etc

      2. Use a ground lifter from each power supply to the chassis

      3. Make sure the Input sockets are right next to each other

      4. ONLY CONNECT THE TWO CHANNELS IN ONE PLACE AND THAT IS AT THE INPUTs

      I hope this helps

      Regards

      Andrew

  20. Cuong Nguyen says:

    Hi Andrew,
    Thanks for you suggests,
    Please explain to me about Common Impedance Coupling Issue. I do not yet completely understand it.
    1. The conditions to occur the issue? The mistakes of wiring/grounding on PCB layout which will happen the issue?
    2. The optimal ways to overcome and prevent.

    Thanks & regards,

  21. Raj says:

    Hi Bonsai,
    I was wondering what fault condition in the transformer would put a non-zero potential at the 0V point?
    I have a active filter which has feeds two amps. I do not connect the 0V to the safety earth (and chassis) in the amps. I connect the 0V to the safety earth (and chassis) only in the active filter. I realize that the fault currents in the RCA interconnects would be disastrous, but was wondering how such a situation would arise.

    • Bonsai says:

      Hello Raj

      I would need to know how you are measuring that to comment. how are you connecting your meter?

  22. Raj says:

    Hi Bonsai,
    Sorry for the confusing question. I do not have a fault situation, but was wondering if my approach is dangerous? I hope my rephrased question make sense.
    I have an active filter which has feeds two amps. I do not connect the 0V to the safety earth (and chassis) in the amps. I connect the 0V to the safety earth (and chassis) only in the active filter. I realize that the fault currents in the RCA interconnect cables would be disastrous, but was wondering how such a situation would arise.
    I was wondering what fault condition in the transformer would put a non-zero potential at the 0V point? Or what fault condition in the amp would do the same?

    • Bonsai says:

      Hello Raj,
      Sorry for my very late reply.

      I would not do this – it is not safe.

      If you have a earth loop, then use a ground lifter in the power amp which will essentially accomplish what you are doing, but give you a path to safety ground (ie safety earth) in the event of a catastrophic fault inside the power amp.

      Seepage 25 in the Ground Loop presentation on how to do this. Importantly:-

      1. The amplifier chassis is connected directly to safety earth and to the one end of the ground lifter bridge rectifier
      2. The amplifier 0V is connected to the other end of the ground lifter bridge rectifier
      3. I recommend you use a bridge rectifier rated at 35 A or greater with a surge capability of 150A or more

  23. Andre Routh says:

    Dear Bonsai,

    Thank you for this mine of information. Whenever I’ve built a preamp, amp, or PSU and the thing hasn’t hummed horribly, I leave it alone without really wondering why it is quiet. Conversely, when dealing with a hummer, I try star grounding, bundling wires, separating wires, etc. I keep going until suddenly the hum is gone or dropped to an acceptable level (no, don’t ask me what that is in dB).

    I’ll apply your techniques on my next build and might achieve your Holy Grail 100db!

    Best Regards,

    Andre Routh, Medord, NJ

  24. Grimrian says:

    Hello Andrew,

    thanks for an informative website, which is of great interest! I have also reverted to my original field of audio engineering, and build amplifiers for fun. Good way to spend time during lockdowns! Anyway, you recommend bypass caps at various locations in a power amplifier construction, but perhaps a few words on capacitor type? I suppose ceramics are best, but the choice of dielectric may have some impact on the result. Also make it easier to choose from the thousands of hits on Mouser. What capacitor/dielectric would you suggest?

    • Bonsai says:

      Thanks for the question Grimrian.
      Probably the best way of looking at this is to think about what’s going on in an amplifier PSU. At very low frequncies (say < few hundred Hz), the resevoir capacitors store all of the energy requirements. In a good 100 W commercial amp, these will typically be between 12000 uF and 22000 uF on each rail. On the amplifier modules, local bulk decoupling is provided and in a power amplifier this is typically 220uF to 1000uF on each rail. On big amplifiers, there may be 4 or 5 sets of these caps on each rail. The thinking here is that the local bulk decoupling provides a lower PSU source impedance than just relying on the resevoir capacitors given wire inductances and wireing resistances. Large resevoir capacitors typically feature Equivlent Series Resistances (ESR) of 10-20 milli-Ohms. At the frequency that the total wiring inductance and resistance from the PSU to the amp module becomes greater than this, local bulk decoupling capacitors will help to reduce it again, or at least prevent it from increasing rapdidly. This 'transition' frequency is unlikely to be less than a few kHz in a practical amplifier. The bulk decouplinig capacitor capacitors are always aluminium electrotlyic types rated at least 20% greater than the worst case max supply voltage of the amplifier. After this, on the amplifier module, a combination of 1uF and 0.01uF capaciors are ideally required. If you look at the impeadance plot of a capacitor, it decreases as frequency increases, and at a certain point, starts to increase again. Before the minima, the overall impedance characteristic is capacitive, and above the minima, it is inductive. The increase in the capacitance impedance takes place because the device lead inductance begins to dominate at HF.So, in an ideal case, you want high value resevoir capacitors placed directly after the rectifiers. On the amplifier modules, you will need local bulk decoupling capacitors of between 220uF to 1000uF. Then, dotted around the board, 1uF, 0.01uF and 0.01 uF capacitors. I use 100V X7R ceramic 1206 SMD capacitors in my commercial amplifiers (see http://www.ovationhifidelity.com). This approach ensuires the PSU impedance at a system level is low and critically, capacitive out to HF and is resonably flat.

      Obviosly, good PCB and wiring layout practice is mandatory – capacitors won’t solve problems caused by sloppy layout – it will only improve a good layout.

  25. Tamas Oszvald says:

    Hello Andrew, I have the problem, that I dont know where is the correct place for a HBR resistor in the Renardson amp MJR7-MK5. I build it, but I am not really happy with the hum. It is between Rin and the first 470p, or between the first and second 470p, or between the second 470p and the 1,5Ohm Resistor?
    I beg your pardon for my poor english!

    Best Regards,

    Tamas Oszvald, Hungary

  26. Chris says:

    Excellent article (seen on DIY audio also).

    For signal input ground capacitor to chassis – other than ceramic is there a particular type of capacitor recommended? A part number would be great to see what is locally available to match it.

    Same question if using a ground loop breaker with parallel resistor (10R) and capacitor (100nF) – what type of ceramic capacitor is recommended?

    Unfortunately some of us DIY amp builders know how to follow a schematic, but, not the finer details of exactly what to use for best outcome.

  27. Eric says:

    Dear Bonsai,

    thank you very much for your excellent, very detailed Ground Loops slides!

    I am in the process of building a preamp based on Rod Elliotts P88 using his P05 +/- 15v PSU and I just came across your invaluable information.

    I have built the modules on Veroboard and have tried to incorporate your hum removal suggestions. Due to the fact the boards are Veroboard (built and working) I could not implement your suggestions exactly, but the preamp does seem very quiet (on the ‘scope, not wired yet for output).

    I would appreciate your comments/suggestions regarding what I have done and propose to do:

    1. I could not incorporate your PSU T connections, but I moved the ground connection from the edge of the board to the junction of the second set of filter capacitors, completely soldered over the 2 Veroboard tracks I used for 0v.

    2. I connected this 0v point to the 0v tracks on the preamp, close to the volume pot ground (as suggested by a DIYAudio member).

    3. I completely soldered together the 2 0v tracks again.

    4. I am using this preamp 0v connection as the star to connect to chassis ground/PE (via a 15R resistor and parallel 100nF ceramic ), an RIAA preamp, possibly 2, and input and output socket grounds.

    I would appreciate any comments/suggestions on what I have done so far.

    Here is where I am uncertain how to proceed:

    I intended to connect all the phono socket grounds together with a 16 AWG wire and connect this bus back to the star 0v on the preamp with a single wire. Is this correct?

    Should I separate the grounds for input/output sockets?

    I then intend to wire the inputs/outputs using screened cable, with the screen connected only at the phono sockets.

    Is my ground loop breaker resistor/capacitor to the chassis/PE star OK, or should I incorporate it into the 0v line from PSU to preamp as you show it?

    Thank you for any assistance, it is much appreciated!

    • Bonsai says:

      Hello Eric, thanks for your kind comments about the presentation.

      It is quite difficult to answer this without any pictures, but I will try below:-

      1. For your power supply, where the two filter capacitors meet, bring a short wire (5mm is ok) to a point on your Veroboard. This point is your clean ground. If you connect anything where the two capacitors meet (lets call this ‘dirty ground’) you run the risk of common impedance coupling charging current pulses into your signal ground (aka ‘clean ground’).
      2. Your input signal grounds must connect to all the RCA input signal grounds, and then a single wire must go to your preamp input ground and from thereto the power supply clean ground.
      3. Your output ground must go separately back to the output ground on the board and from there back to the clean ground. Do not tap it off the input signal grounds
      4. Make sure the input wires and the input ground are tightly twisted together to minimise the loop area
      5. Make sure the output signal wire and output ground are tightly twisted together.
      6. Keep the power supply transformer well away from any circuits – radiated mag fields from the transformer will couple into any loop areas you have and cause noise problems.
      7. For your ground lifter, I would take that off the clean ground
      8. make sure there is one and only one connection between the chassis and 0V (clean ground). If you have more than one connection, you will get noise. Make sure the input and output connectors are 100% insulated from the metal chassis
      9. The safety ground from the mains must go from your input power wire to the point on th4e chassis where your ground lifter is connected – do not use a separate connection in a different location.

      Hope this helps – please feel free to ask for more clarification if needed.

      Kind regards

      Andrew

  28. Eric says:

    Hello,

    I would just like to add to my previous comment that there are 10 pairs of input/output RCA phono connectors.

    Also that in point 3., I am referring to the two 0v tracks on the preamp which I used for 0v, They are side by side, and I soldered them together with a complete solder bridge to reduce resistance. (Same on the PSU)

  29. Eric says:

    Hello Andrew,

    thank you for your prompt and detailed reply.

    Here are some photos:

    https://photos.app.goo.gl/uDr393RHwVZ3UARV7

    1. Complete amp showing the wiring so far. The wire from the chassis star point to the preamp board goes via the ground break resistor.

    2. PSU showing the 0v connection (green) now next to the second set of filter capacitors, instead of at the end of the board like the +/-connections. I wanted to try to emulate your “T” configuration, but you say it’s a bad move.

    3. The preamp board with the 0v star point connected to PSU 0v, chassis star via ground break and ground wires to go to input grounds and RIAA preamp. The output connector grounds are also connected here.

    I am not a great circuit layout person, so the layouts basically follow the schematics. This means that the +/- and 0v tracks on the preamp go from one end of the board to the other. This may not be a good thing to do?

    To your answers:

    1. The power supply is regulated and the original 0v connection point on the edge of the board is about 4 cm from the second set of filter capacitors, past the regulators and associated components, that’s why I moved it closer to the capacitors to try to emulate your “T” connection. Should I take the clean 0v from somewhere along that track or at the end? The 0v track is 2 tracks soldered completely together to lower impedance.

    2 + 3, The preamp board has inputs on the left and outputs on the right side. I originally had planned to use 0v/signal ground connections at these ends, but changed to a star connection at the volume control ground, to only have one star ground point on the board, after reading lots about grounding.. The input and output ends of the board are 18 cm apart. If I use the ground connections both at the input and output ends, doesn’t this create a loop problem?

    Also, am I misunderstanding your comment “… single wire must go to your preamp input ground and from there to power supply clean ground”. Do you mean two connections for input and for output grounds, the second to PSU clean ground?

    Or should there be only one clean ground connection to the preamp board and if so where on the preamp is best?

    4 + 5. I was planning to use screened cable for inputs and outputs, with the screen connected only at the RCA connector end, with signal grounds all bussed together and returned via a single wire to 0v/signal ground, as you mentioned. Is this not OK? I don’t know how it would work with twisted pairs, would I leave the ground wire disconnected at one end, otherwise there would be multiple ground connections?

    6. I hope the transformer location will be OK, I have left the wires long so far to try rotating it for best null, as you suggest in your presentation. The power switch has not arrived yet….

    7, 8, 9. All OK.

    With my star 0v connection on the preamp board I was using the suggestions of Fast Eddie on DIYAudio and it’s the same as in Dave Davenport’s article on Grounding, Figure 3.2-9, the right hand side example:
    https://www.diyaudio.com/community/threads/audio-component-grounding-and-interconnection.163575/

    I think your suggestions are the left hand side option?

    Sorry for so many questions! I thought I had it all sorted out, but I appreciate your help very much! Looking forward to getting it finally wired up!

    Thank you!!!

Leave a Reply

Your email address will not be published. Required fields are marked *