Rick, I don't think there is any sonic advantage to using a higher voltage cap here, and it might even be detrimental. There is definitely a change in sonics with different mfd value caps. After all my experiments, I've concluded that somewhere between 1.0 (Mauro's original choice) and 1.5uF is best. I think Andrew's previous posts about his feedback calculations bear that out, although, if I understand it correctly, his results point to an even smaller input cap or a larger feedback cap.
I have never tried running the amps without the cap, and never plan to do so. Although I use a passive preamp with autoformer volume control which should block any DC (there is no output cap), I can sometimes hear crackling when turning the control knob. I suspect the contacts are not perfectly clean, and I bet that crackling contains a DC component which I do not wish to pass along to the amps. I just don't think the sonic benefit is worth the risk of removing that cap and blowing stuff up. People are so very paranoid about caps in the signal path, but I kind of like being able to somewhat tailor the sound to my liking and the rest of the system. I seek accuracy but also musicality; absolute purity of signal is not always the most pleasing sound. As I learned from trying the Sonicap, you don't need to spend a fortune to get good results.
Peace,
Tom E
I have never tried running the amps without the cap, and never plan to do so. Although I use a passive preamp with autoformer volume control which should block any DC (there is no output cap), I can sometimes hear crackling when turning the control knob. I suspect the contacts are not perfectly clean, and I bet that crackling contains a DC component which I do not wish to pass along to the amps. I just don't think the sonic benefit is worth the risk of removing that cap and blowing stuff up. People are so very paranoid about caps in the signal path, but I kind of like being able to somewhat tailor the sound to my liking and the rest of the system. I seek accuracy but also musicality; absolute purity of signal is not always the most pleasing sound. As I learned from trying the Sonicap, you don't need to spend a fortune to get good results.
Peace,
Tom E
Hey guys,
I have a question for you guys. Some time ago I've build a My Ref RevC amp and I used it hooked up directly to my laptop for some time with great pleasure. Last weekend I finished my Mesmerize DCB1 preamp and I connected it to the My Ref. A very loud humm on both speakers was the result, even when the preamp was still off. Connected to another amplifier the Mesmerize seems fine, which leads me to believe it has something to do with the My Ref. When I only connect only one channel preamp to amp, there is no hum.
I've attached a pic of the inside of the amp. It's a dual mono setup, with both transformers housed inside a seperate wooden box. There is no connection to mains ground and also there is no cap connected at the input, as is included in the ultimate BOM.
Does anyone have an idea of what might be wrong here?
Thanks!
Erik
I have a question for you guys. Some time ago I've build a My Ref RevC amp and I used it hooked up directly to my laptop for some time with great pleasure. Last weekend I finished my Mesmerize DCB1 preamp and I connected it to the My Ref. A very loud humm on both speakers was the result, even when the preamp was still off. Connected to another amplifier the Mesmerize seems fine, which leads me to believe it has something to do with the My Ref. When I only connect only one channel preamp to amp, there is no hum.
I've attached a pic of the inside of the amp. It's a dual mono setup, with both transformers housed inside a seperate wooden box. There is no connection to mains ground and also there is no cap connected at the input, as is included in the ultimate BOM.
Does anyone have an idea of what might be wrong here?
Thanks!
Erik
Attachments
I see untwisted sets of flow and return wiring.
The speaker flow and return have enormous loop area directly above the amp PCB and one or the speaker pair loop encloses the input cable pair, (almost current transformer).
The AC power into your PCBs is also untwisted but this time much smaller loop area.
What is the green/yellow alongside the AC?
Do you have any safety precautions/procedures/methods in place while you have exposed conductive parts outside your wood casing?
The speaker flow and return have enormous loop area directly above the amp PCB and one or the speaker pair loop encloses the input cable pair, (almost current transformer).
The AC power into your PCBs is also untwisted but this time much smaller loop area.
What is the green/yellow alongside the AC?
Do you have any safety precautions/procedures/methods in place while you have exposed conductive parts outside your wood casing?
Hi Andrew,
Thanks for your answer. I will improve the wiring going to the speakers.
The green/yellow is one of the four wires coming out of the transformer. It is combined with the blue wire.
Currently I have no safety precautions installed. The only conductive part outside the wooden case are the heat sinks, but they are not connected to anything conducting. And then there are the rca and speaker terminals of course. However, reading through a topic about star grounding the other day taught me to take precautions. I assume that is what you implicitly mean by asking that question?
Would connecting the heat sink to a star ground and then connect it to mains ground for each channel do the trick (see picture)? What would be a good point to connect it to on the pcb?Regards,
Erik
Attachments
Hi,
if there is any possibility that anyone could touch any of the exposed metal parts while a mains fault is occurring then you must have a serious re-think.
Could a dropped offcut of resistor leg fall into the mains power box?
Could a pet or child touch something while you can't see them?
Can you post a wiring diagram of all the grounds and Zero volts and signal returns and how they all interconnect?
DO NOT put in that 6r "break" resistor.
That audio ground to chassis connection must be capable of passing Fault Current until the fuse ruptures and arc extinguishes or the Distribution board MCB opens and arc extinguishes.
That Fault Current can be many thousands of amperes. 6ohms in that route may prevent the fuse from rupturing for many minutes or even hours and in the meantime the exposed conductive components could be at or near mains voltage,
if there is any possibility that anyone could touch any of the exposed metal parts while a mains fault is occurring then you must have a serious re-think.
Could a dropped offcut of resistor leg fall into the mains power box?
Could a pet or child touch something while you can't see them?
Can you post a wiring diagram of all the grounds and Zero volts and signal returns and how they all interconnect?
DO NOT put in that 6r "break" resistor.
That audio ground to chassis connection must be capable of passing Fault Current until the fuse ruptures and arc extinguishes or the Distribution board MCB opens and arc extinguishes.
That Fault Current can be many thousands of amperes. 6ohms in that route may prevent the fuse from rupturing for many minutes or even hours and in the meantime the exposed conductive components could be at or near mains voltage,
Last edited:
hi Udailey,
I've attached a picture of my test rig for the DCB1. Can you to me a bit more of what you exactly mean? I've been studying pictures of completed mesmerizes by other users and I can't find significant differences.
Regards,
Erik
Attachments
stereo signal with common return
Something I cannot get my head around regarding running a three wire two channel signal when the signals share a common third (return) wire.
How can we minimise the loop area of each channel's flow and return, without compromising the crosstalk/isolation between the two channels?
I bump into this in most of my builds (except monoblocks) and I can see the problem occurring four times in the pic of the previous post.
Something I cannot get my head around regarding running a three wire two channel signal when the signals share a common third (return) wire.
How can we minimise the loop area of each channel's flow and return, without compromising the crosstalk/isolation between the two channels?
I bump into this in most of my builds (except monoblocks) and I can see the problem occurring four times in the pic of the previous post.
Last edited:
A very loud buzzing is often the fault of a bad signal ground. You may have a bad solder joint to ground in one of your components but good signal grounding in the other component saves you when you plug into it - just a possibility. You have to find where you are not grounding the signal. I dont see any obvious signal grounding problems, however so I could be wrong. You are trying to find the fault in one of three devices and these pics do not show us the problem.
I don't have any general solution, but for the MyRef, I have done the following with no issues so far:
1) Run single-core shielded wires from each monoblock's input connector to the RCA connector block.
2) Connect the shield of each of these wires to the 'lifted' signal ground at each monoblock.
3) Connect both shields to the shield terminal of the RCA connector block.
4) Do not connect the shield terminal at the RCA to the chassis ground or any star ground - i.e. leave it isolated/floating from the chassis.
This works for me, but there is still scope for a ground loop between the two shields of the signal cables between the source and the MyRef if they're separate left and right cables (which many audiophiles seem to believe reduces crosstalk).
The way to minimize the enclosed area in this ground loop is to use two separate single-channel shielded cables that are physically bonded together as a dual moulded cable for almost their entire length (this is the default configuration of most garden-variety line cables. Ironically, the cheap stuff works better than the audiophile-grade stuff here).
It's still not an issue with the method I proposed, right?
There should be no ground loop on the PCB for the lifted ground across the 2 channels (easy to ensure). The shared return is connected to the lifted/isolated ground. It connects to the chassis RCA block with maybe a 2-core shielded cable (with crosstalk issues) or a dual-bonded shielded single cable. Source to RCA connector is the same dual-bonded shielded cable. There are minor ground loops between shields, as before.
Vaguely (un)related: in the pic of the preamp (?) board shown in post #1147, I see unshielded teflon-insulated runs between the input RCA and the board, as well as the board and output RCA. They're short runs, but they're still going have pickup and radiation. Using dual-core shielded cables or dual single-shielded cables is nevertheless the better option for both input and output, even if it isn't Teflon insulated.
I'm guessing that the output return is isolated from the input signal ground, probably with a lift resistor for the input signal ground. Otherwise, there's still a possibility of a large ground loop through the (non-isolated) chassis grounds and the earth leads of the power amp (downstream) and the source (upstream).
I'm guessing that the output return is isolated from the input signal ground, probably with a lift resistor for the input signal ground. Otherwise, there's still a possibility of a large ground loop through the (non-isolated) chassis grounds and the earth leads of the power amp (downstream) and the source (upstream).
I can confirm that there is no "loop breaker" between the signal ground and the output ground.
Similarly the two ground runs to the volume pot are also on this same common grounding trace.
Thanks for conforming that you would still use close coupled coax for all the short runs.
That's a practice I gave up more than two decades ago, maybe time to revisit. Strip the outer insulation and wrap a copper binding wire around the two shields? food for thought?
Similarly the two ground runs to the volume pot are also on this same common grounding trace.
Thanks for conforming that you would still use close coupled coax for all the short runs.
That's a practice I gave up more than two decades ago, maybe time to revisit. Strip the outer insulation and wrap a copper binding wire around the two shields? food for thought?
Last edited:
Hey All,
Can someone tell me if it is worth the effort/cost to buy/build a metal can or box to isolate the toroids in a Ref-C build to minimize interference? If yes, what is the metal of choice? If no, is there a recommended minimum distance between the board and the transformer?
Thanks, Bob M.
Can someone tell me if it is worth the effort/cost to buy/build a metal can or box to isolate the toroids in a Ref-C build to minimize interference? If yes, what is the metal of choice? If no, is there a recommended minimum distance between the board and the transformer?
Thanks, Bob M.
leave the AC leads from/to the transformer long.
install and get working.
Rotate the transformer around it's (loose) mounting bolt while measuring the noise at the outputs of the amplifiers.
You may find a NULL that minimises the noise in one channel or the other. If you are lucky you may find a rotation that gets both channels almost silent.
Once this is completed, trim all the AC leads to minimum length, without them getting too close to signal lines.
install and get working.
Rotate the transformer around it's (loose) mounting bolt while measuring the noise at the outputs of the amplifiers.
You may find a NULL that minimises the noise in one channel or the other. If you are lucky you may find a rotation that gets both channels almost silent.
Once this is completed, trim all the AC leads to minimum length, without them getting too close to signal lines.
Hi Bob,
I build two RevC's with no separate metal enclosure for the trafos, both are dead silent. Need to be careful keeping the AC line away from the amp boards and input/outputs. Also try to locate the trafos as far away in the enclosure from the amp boards as possible. All in all not a big deal.
PJN
I build two RevC's with no separate metal enclosure for the trafos, both are dead silent. Need to be careful keeping the AC line away from the amp boards and input/outputs. Also try to locate the trafos as far away in the enclosure from the amp boards as possible. All in all not a big deal.
PJN
I found that hum is not an issue even with a single EI-core trafo for both the MyRefs - mine, however, had a wide copper-foil 'waist-band' that absorbs the magnetic flux leakage from the EI-core by creating a local ground loop around it. I don't think that Toroidals will have sufficient leakage flux to cause hum pickup problems with a correctly grounded MyRef that is (say) about 6 inches away. The signal ground plane is also isolated from the power ground islands by a 1-ohm isolator, which seems to work well in practice.