Kevin Haskins said:That is a great drawing. A couple notes:
#1. Even though the softstart docs show the AC power ground going to the chassis Hypex suggest floating the AC power ground from the chassis. The only point that connects to the chassis is Pin number one on the XLR input. It should be firmly tied to ground at the XLR input.
Hi,
I don't think that's correct at all. They recommend floating the PSU common from the chassis, not the mains ground. Also where do you plan on grounding your static shield? In order to constitute as being double insulated both the mains earth and static shield should tie to the chassis, unless you take other measures which are no doubt more involved.
I believe they did recommend not earthing the chassis but in the case of unbalanced inputs only.. I believe there are other ways around that which I find preferable, like leaving the chassis earthed and floating the inputs. Some experimenting would be required to determine the exact solution there, I just went with balanced instead for obvious reasons.
Regards,
Chris
Does that transformer double as a WOMD? Sure looks like it!
Regards,
Sander Sassen
http://www.hardwareanalysis.com
Regards,
Sander Sassen
http://www.hardwareanalysis.com
Kevin Haskins said:It is fairly close...
Dia. 5.7"
Ht. 5"
Kevin,
looks like they turned out much too tall for my enclosures. The tallest parts I could fit in there need to be less than 4" in their shortest dimension.
Bummer, so back to plan B and three separate transformers.
Peter
Ok... our plans changed slightly and I ended up ordering shorter transformers than the sample. They are the same every other way but they are going to be 3.15" tall by 6.85" in diameter.
I don't expect inventory for another 7-8 weeks (probably the end of March or beginning of April).
These include all the mounting hardware, potted center, goss shield, copper shield between primary/secondaries etc... All secondaries the same as the pictured sample. Primaries are 115V/230V and all leads are 8" long.
I don't expect inventory for another 7-8 weeks (probably the end of March or beginning of April).
These include all the mounting hardware, potted center, goss shield, copper shield between primary/secondaries etc... All secondaries the same as the pictured sample. Primaries are 115V/230V and all leads are 8" long.
I'll keep them in mind for my second pair of amps. For the first one, I now have everything on hand except for the enclosures. Those are on a rowing boat somewhere in the middle of the Pacific for the next 4 weeks or more.
I bought transformers from Amveco at Digikey for the lower voltages, and the 800VA 60-0 60-0 Avels from Parts Express for each Monoblock. I'll soon be able to begin assembly of a test rig without chassis. Have to find something for heat sinking for the module as well (old Pentium 2 CPU heat sink should do for testing)
Dies anybody know of a good resource with a clear explanation of what "double insulated construction methods" mean in the context of building a power amp? I think I'll rather ground my chassis before I build some sort of death trap only to follow the UCD instructions to the letter and screw up on one of these rather important details.
Peter
I bought transformers from Amveco at Digikey for the lower voltages, and the 800VA 60-0 60-0 Avels from Parts Express for each Monoblock. I'll soon be able to begin assembly of a test rig without chassis. Have to find something for heat sinking for the module as well (old Pentium 2 CPU heat sink should do for testing)
Dies anybody know of a good resource with a clear explanation of what "double insulated construction methods" mean in the context of building a power amp? I think I'll rather ground my chassis before I build some sort of death trap only to follow the UCD instructions to the letter and screw up on one of these rather important details.
Peter
The reasoning behind single and double insulation is that there should always be one extra line of defense between the user and live voltages when something fails. This on the premise that two unrelated faults do not occur simultaneously.
Counted as a single insulation is:
*One layer of insulating material around a conductor
*3mm of air between an uninsulated live part and any externally accessible part (includes the ground connection). You should be able to guarantee these 3mm even with reasonable force exerted on all internal wiring.
Single insulation is acceptable if the apparatus is connected to safety earth. The second line of defense is the differential circuit breaker.
Counted as a double insulation is:
*Two separate layers of insulating material. If one layer is damaged, there should always be a second one.
*One layer of insulating material plus 3mm of air. Again, you should be able to guarantee this even when tugging at the wiring. If the insulation fails, there is still 3mm of air. If something is severely forced and the wire contacts the chassis, there's still the insulation.
*6mm of air.
You can buy double-insulated wires for the primary wiring. This is recognizable when you cut the wire, you get an inner layer of insulation which is white and another one which is coloured. You'll find that the first layer can separate from the second one so that a superficial cut in the insulation doesn't propagate all the way to the conductor when the cable is bent sharply.
Now, I would like to stress that I don't want to force anyone to disconnect the chassis from mains earth. The point is, I can't recommend separating the audio ground from the chassis ground, because that's a recipe for making a radio receiver. However, if one insists on using unbalanced (RCA) inputs or if the ancillary equipment is sensitive to ground loops, tying the audio ground to mains earth causes hum problems or more subtle sonic issues. In other words, the fact that tying chassis and audio grounds together is the only correct solution implies that earthing the chassis creates ground loops. Correctly designed balanced connections are completely insensitive to ground loops (which is why they're universally used in pro audio although not all pro gear is correctly done, see "pin 1 problem"). One should not go around trying to avoid ground loops, because in any system more complicated than a home stereo, ground loops are a fact of life. Instead, one should design their equipment to be impervious to ground loops.
So if everyone would design their gear according to AES48, all equipment could be earthed with no adverse effects. Until that's the case, there'll always be equipment that'll produce hum when connected to another piece of gear that is grounded.
So, there's the choice:
A) If you want to use RCA inputs, disconnect the mains earth and employ double insulated construction techniques.
B1) Use balanced (XLR) inputs. This allows the whole thing to be earthed unless the ancillary equipment has problems.
B2) Make a "pseudo-differential" RCA input. I still haven't figured out whether or not I should post a detailed description of how to do this, because unless I manage to explain with perfect clarity it's almost certain to generate large volumes of mail.
C) Anthing else (e.g. floating the amps inside a grounded chassis), but then you're on your own if you hear your mobile through the speakers.
Counted as a single insulation is:
*One layer of insulating material around a conductor
*3mm of air between an uninsulated live part and any externally accessible part (includes the ground connection). You should be able to guarantee these 3mm even with reasonable force exerted on all internal wiring.
Single insulation is acceptable if the apparatus is connected to safety earth. The second line of defense is the differential circuit breaker.
Counted as a double insulation is:
*Two separate layers of insulating material. If one layer is damaged, there should always be a second one.
*One layer of insulating material plus 3mm of air. Again, you should be able to guarantee this even when tugging at the wiring. If the insulation fails, there is still 3mm of air. If something is severely forced and the wire contacts the chassis, there's still the insulation.
*6mm of air.
You can buy double-insulated wires for the primary wiring. This is recognizable when you cut the wire, you get an inner layer of insulation which is white and another one which is coloured. You'll find that the first layer can separate from the second one so that a superficial cut in the insulation doesn't propagate all the way to the conductor when the cable is bent sharply.
Now, I would like to stress that I don't want to force anyone to disconnect the chassis from mains earth. The point is, I can't recommend separating the audio ground from the chassis ground, because that's a recipe for making a radio receiver. However, if one insists on using unbalanced (RCA) inputs or if the ancillary equipment is sensitive to ground loops, tying the audio ground to mains earth causes hum problems or more subtle sonic issues. In other words, the fact that tying chassis and audio grounds together is the only correct solution implies that earthing the chassis creates ground loops. Correctly designed balanced connections are completely insensitive to ground loops (which is why they're universally used in pro audio although not all pro gear is correctly done, see "pin 1 problem"). One should not go around trying to avoid ground loops, because in any system more complicated than a home stereo, ground loops are a fact of life. Instead, one should design their equipment to be impervious to ground loops.
So if everyone would design their gear according to AES48, all equipment could be earthed with no adverse effects. Until that's the case, there'll always be equipment that'll produce hum when connected to another piece of gear that is grounded.
So, there's the choice:
A) If you want to use RCA inputs, disconnect the mains earth and employ double insulated construction techniques.
B1) Use balanced (XLR) inputs. This allows the whole thing to be earthed unless the ancillary equipment has problems.
B2) Make a "pseudo-differential" RCA input. I still haven't figured out whether or not I should post a detailed description of how to do this, because unless I manage to explain with perfect clarity it's almost certain to generate large volumes of mail.
C) Anthing else (e.g. floating the amps inside a grounded chassis), but then you're on your own if you hear your mobile through the speakers.
Extra voltages not a big deal
If you just get a toriod with an open core, adding turns for a few extra rails is no big deal. The Avel transformers allow this for example.
Just get your 800VA 60V transformer from parts express.
(I don' t know about there hum, but I've used them in the past, older models, and they were okay.)
Get it, and count the big secondary wire turns. The volts per turn will be 60V/number or turns for one of the secondary windings.)
If its hard to keep track, just count them all and 120/all the turns.) The turns are not that many. You might only need 5 turns or even less to get 12V.
You will want to try to evenly space them for lowest magnetic field.
Avel used to, and still may sell a custom kit, where they wind the primary and give you wire for secondaries, and tape to put over it.
You could use electrical tape, but what they supply will be higher temp.
Best Regards,
MIke
If you just get a toriod with an open core, adding turns for a few extra rails is no big deal. The Avel transformers allow this for example.
Just get your 800VA 60V transformer from parts express.
(I don' t know about there hum, but I've used them in the past, older models, and they were okay.)
Get it, and count the big secondary wire turns. The volts per turn will be 60V/number or turns for one of the secondary windings.)
If its hard to keep track, just count them all and 120/all the turns.) The turns are not that many. You might only need 5 turns or even less to get 12V.
You will want to try to evenly space them for lowest magnetic field.
Avel used to, and still may sell a custom kit, where they wind the primary and give you wire for secondaries, and tape to put over it.
You could use electrical tape, but what they supply will be higher temp.
Best Regards,
MIke
UCD700 Layout and wiring
This is actually more of a bump for another thread I started on ideas for layout and wiring of the UCD700. No replies yet, so hopefully someone who posted here will be interested.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=88255
This is actually more of a bump for another thread I started on ideas for layout and wiring of the UCD700. No replies yet, so hopefully someone who posted here will be interested.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=88255
Perhaps some of you will be interested in knowing that Coldamp has developed a special version of our SPS80 1KW switching power supply with +/-85V outputs, should suit UcD700 as well.
Bruno Putzeys said:One should not go around trying to avoid ground loops, because in any system more complicated than a home stereo, ground loops are a fact of life. Instead, one should design their equipment to be impervious to ground loops.
So a passive pre is out of the question because of having an unbalanced output? The only solution will be a transformer with balanced outputs.
In theory it is all very nice to have everything balanced but as long as the average gear you buy in the shop has RCA outputs, well there is not much of an option here. Better isolate your PSU to be able to have 3Kv isolation between mains parts and chassis ground.
- Status
- Not open for further replies.
- Home
- Amplifiers
- Class D
- Optimal transformer for UCD700