The amplifier is already capacitively loaded - the output capacitor is 680nF. As long as what you're attaching is lower than this (order of magnitude) you're safe. Especially if you're using speaker cables. The impedance of even a short speaker cable is high enough to "decouple" the load from the amp.Dave S said:
I don't think you should worry.
Re: System 0V & Chassis Earth
In balanced connections, the chassis are connected through the cable. In unbalanced connections, this is not healthy (hum), so there you would have the audio ground floating with respect to chassis. Unfortunately, floating grounds are an EMI hazard (susceptibility, not emission). You then see folks adding small caps between chassis and signal ground at the input connectors in order to make an HF connection. In doing so they create a brilliant hf current loop inside the device between the internal "star" and all the connectors. Such contraptions make my stomach turn. With simple audio setups, one may have some sympathy for the seeming simplicity of unbalanced operation but once things get complicated they get ugly too.
Balanced audio equipment is normally not designed to have a floating ground (it is tied to chassis), simply because ground is not used as part of the signal. The sound quality advantage of correctly designed balanced connections is often underestimated (possibly due to the relative rarity of correctly designed ones).
Of course, some people do still use floating grounds combined with balanced wiring. Thus they invite EMI ingress problems. Their choice, not mine.
FuriousD said:
In balanced connections, the chassis are connected through the cable. In unbalanced connections, this is not healthy (hum), so there you would have the audio ground floating with respect to chassis. Unfortunately, floating grounds are an EMI hazard (susceptibility, not emission). You then see folks adding small caps between chassis and signal ground at the input connectors in order to make an HF connection. In doing so they create a brilliant hf current loop inside the device between the internal "star" and all the connectors. Such contraptions make my stomach turn. With simple audio setups, one may have some sympathy for the seeming simplicity of unbalanced operation but once things get complicated they get ugly too.
Balanced audio equipment is normally not designed to have a floating ground (it is tied to chassis), simply because ground is not used as part of the signal. The sound quality advantage of correctly designed balanced connections is often underestimated (possibly due to the relative rarity of correctly designed ones).
Of course, some people do still use floating grounds combined with balanced wiring. Thus they invite EMI ingress problems. Their choice, not mine.
Sometimes it's easier to ask than to check or simulate;
I can understand that the build upp on the psu can happen when there is some dc at the output. And that it will happen on the opposit supply. Caused by the conduction / current try to continue flowing in the output coil.
Serie diode with the outputs sorse side on positive psu and drain side on neg psu. will then (?) make the voltage at the coil conection go almost
(useless)
But storing/losing some energy in a zobel ( R with series C) across each switching device it should bee more tolerant to dc ofsett ( energy stored/energy consumed) ie reverse conduction of energy redused by stored energy-losses, may bee difficoult for the full range of output. For a smal amount the losses is not so big either. The tolerance for dc I GUESS will bee higher without pumping. But has some drawbacks regarding zwitching and distortion. Somtimes in simulation i find the switching performance kind of better, the voltage at the switched side of coil do not swing to the opposit end when not conduckting. And the voltage feeding the coil have less ripple hence the output ripple/noise will also bee redused by the same amount.
Losses however.... a small compromise maybee ?
( No offence taken!)
On the UCD 400 pics there seem to bee a smal open space so that double outputs could bee monted, is it so? ( 8psc dvc of 8+8 Ohm [6.8 Ohm dc] for my bottom and i'm just starting to think about it.... By now my home made with a huge amount of output dewices is feeding them )
And is the UCD 400 available already?
Can it bee monted in a 1u h cabinett ie 38mm high no more?
I can understand that the build upp on the psu can happen when there is some dc at the output. And that it will happen on the opposit supply. Caused by the conduction / current try to continue flowing in the output coil.
Serie diode with the outputs sorse side on positive psu and drain side on neg psu. will then (?) make the voltage at the coil conection go almost
(useless)But storing/losing some energy in a zobel ( R with series C) across each switching device it should bee more tolerant to dc ofsett ( energy stored/energy consumed) ie reverse conduction of energy redused by stored energy-losses, may bee difficoult for the full range of output. For a smal amount the losses is not so big either. The tolerance for dc I GUESS will bee higher without pumping. But has some drawbacks regarding zwitching and distortion. Somtimes in simulation i find the switching performance kind of better, the voltage at the switched side of coil do not swing to the opposit end when not conduckting. And the voltage feeding the coil have less ripple hence the output ripple/noise will also bee redused by the same amount.
Losses however.... a small compromise maybee ?
( No offence taken!)
On the UCD 400 pics there seem to bee a smal open space so that double outputs could bee monted, is it so? ( 8psc dvc of 8+8 Ohm [6.8 Ohm dc] for my bottom and i'm just starting to think about it.... By now my home made with a huge amount of output dewices is feeding them )
And is the UCD 400 available already?
Can it bee monted in a 1u h cabinett ie 38mm high no more?
Konrad said:
What a complicated way of losing energy! Life's quite simple. If you have pumping, providing extra energy loss will reduce the problem. Whether you do this by making the snubbers heavier than necessary or by simply placing a ballast resistor across the power rails is not instrumental... Just try not to have DC at the output. Nobody needs DC at a speaker output anyway
Cheers,
Bruno
Konrad,
The UcD400 is in order by our PCB manufacture and assembler. In about 6 weeks (perhaps sooner?) we will have them on stock.
What do you mean with this question?
The connectors are as follows from left to right;
4 input connector
1 - +input
2 - ground
3 - -input
4 - amp_on
2 pin power connector
1 - power out hot
2 - power out cold
3 pin power connector
1 - +powersupply
2 - ground
3 - -powersupply
Indeed it can be mounted in 1 HE case
Regards,
Jan-Peter
www.hypex.nl
The UcD400 is in order by our PCB manufacture and assembler. In about 6 weeks (perhaps sooner?) we will have them on stock.
What do you mean with this question?
The connectors are as follows from left to right;
4 input connector
1 - +input
2 - ground
3 - -input
4 - amp_on
2 pin power connector
1 - power out hot
2 - power out cold
3 pin power connector
1 - +powersupply
2 - ground
3 - -powersupply
Indeed it can be mounted in 1 HE case
Regards,
Jan-Peter
www.hypex.nl
Peranders,
Not exactly, we have a symmetrical feedback straight from the outputterminals. So the cold output is nearly ground
Regards,
Jan-Peter
www.hypex.nl
Not exactly, we have a symmetrical feedback straight from the outputterminals. So the cold output is nearly ground
Regards,
Jan-Peter
www.hypex.nl
I meant switching devices, but that is minor concern. To reform the Q i was intrested in the available current output.
So there is a chance for somthing spesial under the x-mas tree i guess.
if not
il have to make somthing else.Abt those losses, i dont actually mind where they are as long as the contribution to better performance is good.
and not too big either 
Konrad,
Ah more Powerfets? This will not be possible, because extra powerfets will increase the switching losses. And dergreaded the performance. Don't forget we are talking here about a switching amp who switchs at 450kHz and in about 20-50nS from zero to +60VDC. ( it can be even faster, I will check this tomorrow)
It's not a liniear amplifier
I don't know when you have Christmas ( ) but within 1 to 1.5 month we have the UcD400 on stock!
Regards,
Jan-Peter
www.hypex.nl
Ah more Powerfets? This will not be possible, because extra powerfets will increase the switching losses. And dergreaded the performance. Don't forget we are talking here about a switching amp who switchs at 450kHz and in about 20-50nS from zero to +60VDC. ( it can be even faster, I will check this tomorrow)
It's not a liniear amplifier
I don't know when you have Christmas (
) but within 1 to 1.5 month we have the UcD400 on stock!Regards,
Jan-Peter
www.hypex.nl
wiring
1. The screen may act as an antenna, picking up voltages by influence at high impedances, or currents by induction at lower impedances. Well known as screen current(?, german: Mantelstrom) in the RF-area.
2. Equalising currents may reach levels in the upper mA-range, if the mains wiring is distributed (not equalised loads) and if there are loads connected like PCs. Try to measure the voltage between your satellite receiver's case and the coax cable shield. Such voltages may be dangerous to the audio equipment's inputs.
I believe, it is better to have an extra earth cable, which should be connected to the mostly available earth screw at the metal cases (star ground). I would use one-end-grounding of the screen with stationary equipment. It may be not useful with stage equipment, because of the wiring effort.
BTW: not grounded cases may act as an antenna theirselves, influencing the circuits inside, at least in common mode.
Excuse me the off topic issue again, I believe, it is worth to be told for keeping all the wonderful amp's capabilities.
Regards, Timo
This addresses two problems:
1. The screen may act as an antenna, picking up voltages by influence at high impedances, or currents by induction at lower impedances. Well known as screen current(?, german: Mantelstrom) in the RF-area.
2. Equalising currents may reach levels in the upper mA-range, if the mains wiring is distributed (not equalised loads) and if there are loads connected like PCs. Try to measure the voltage between your satellite receiver's case and the coax cable shield. Such voltages may be dangerous to the audio equipment's inputs.
I believe, it is better to have an extra earth cable, which should be connected to the mostly available earth screw at the metal cases (star ground). I would use one-end-grounding of the screen with stationary equipment. It may be not useful with stage equipment, because of the wiring effort.
BTW: not grounded cases may act as an antenna theirselves, influencing the circuits inside, at least in common mode.
Excuse me the off topic issue again, I believe, it is worth to be told for keeping all the wonderful amp's capabilities.
Regards, Timo
Since we are so far OT, perhaps I can ask a fairly basic question:
Based on what is discussed so far on the balanced idea: this seems to imply that the correct design for a balanced interconnect would therefore be a twisted pair for signal wire, with a shield around both of these. The shield should be connected to earth at both ends?
Has anyone experimented with any particularly successful DIY designs for balanced cables, I see mostly unbalanced designs out there?
Based on what is discussed so far on the balanced idea: this seems to imply that the correct design for a balanced interconnect would therefore be a twisted pair for signal wire, with a shield around both of these. The shield should be connected to earth at both ends?
Has anyone experimented with any particularly successful DIY designs for balanced cables, I see mostly unbalanced designs out there?
Bruno,
This is certainly quite a minefield. Unfortunately for those of us who have to make products that will be used in just about every Hi-Fi configuration, the Chassis connected earth is just not a usable reality. Even in a balanced system, there is usually some 0v referenced section and if this is connected to Chassis earth (Even if its only the Unbalanced outputs), then you introduce every switch mode PSU's unwanted garbage into your audio reference point, and its certainly very audible and will only become a problem that worsens as new regulations push more PSU's towards SMPS
To say that its just a case of Balanced = Connected and Unbalanced = Floating, just is not practical as many circuits may have a differential input but then use a single ended amplification stage for one of many reasons (although these are mostly cost of course).
The HF capacitor connection is usually a trade off to try to avoid RF pickup at a high impedance input (that todays great op-amp designs will usually fold down into the audio band for you) whilst trying to avoid a direct connection and hence pollution of the audio reference point (Single ended). I agree this is not ideal.
Most Hi-Fi manufacturers do it their own way. There is no unified earthing policy, and certainly not in the professional field. This leads to trying to deal with all manner of earthing connections at both inputs and outputs. Also a wide range of true and Pseudo differential outputs does not help this either. That said, what is commonly seen amoungst most manufacturers is a floating system 0v. with the Chassis's connected via the Mains earth connection. UK, Euro and US plugs all allow for this. The balanced cable shield connection is then dependant on the input stage design and setup. A bit of each to his own...
Theres alot more to this and I think it is worth exploring on this thread as it has great relavence to the sound quality achieveablewith the UCD modules.
F.
Hi,
Rane Audio has a set of articles which cover the interconnection schemes extensively.
Here's the link for anyone who cares,
http://www.rane.com/library.html
Look under "Wiring, Interconnection & Grounding"
Note #110 and #151 "Grounding and Shielding Audio Devices"
Cheers,
Chris
Rane Audio has a set of articles which cover the interconnection schemes extensively.
Here's the link for anyone who cares,
http://www.rane.com/library.html
Look under "Wiring, Interconnection & Grounding"
Note #110 and #151 "Grounding and Shielding Audio Devices"
Cheers,
Chris
Preamp Match for UCD180
I currently use an old NAD 116 Preamp with my Music Fidelity Typhoon Power amps (also old ), I am going to replace the Typhoons with amps using the UCD180s and as I was planning to keep the NAD as Pre for now. The NAD has Balanced outputs available, from earlier conversations it seems that it would be best to use the balanced connections to the UCD180's, is that right or have I misunderstood something (quite likely !).
I am pretty new to this DIY stuff but isn't there a better way of controlling volume than using a preamp for digital power amps? I can see that something is needed to switch between inputs but can't all of the volume control be done in the digital domain somehow?
I currently use an old NAD 116 Preamp with my Music Fidelity Typhoon Power amps (also old
), I am going to replace the Typhoons with amps using the UCD180s and as I was planning to keep the NAD as Pre for now. The NAD has Balanced outputs available, from earlier conversations it seems that it would be best to use the balanced connections to the UCD180's, is that right or have I misunderstood something (quite likely !).I am pretty new to this DIY stuff but isn't there a better way of controlling volume than using a preamp for digital power amps? I can see that something is needed to switch between inputs but can't all of the volume control be done in the digital domain somehow?
Hi,
The UCD isn't a "digital" amp as you seem to be thinking, it's entirely analog, so you can't just throw in a digital volume without degrading the sound quality of it.
I would be using an Audigy sound card as a pre-amp, sadly the outputs are disgusting on it, I can make an adapter but even that is far from ideal, might have to explore that a bit, but a sound card is my goal for a pre amp.
I've a cheap adapter made up with some 12 gauge monster cable that's over a dozen feet long and the audigy is able to drive it flawlessly though, no noise on it at all (audible) until I turn the computer off, then a 60hz hum takes over with a vengeance.
Regards,
Chris
The UCD isn't a "digital" amp as you seem to be thinking, it's entirely analog, so you can't just throw in a digital volume without degrading the sound quality of it.
I would be using an Audigy sound card as a pre-amp, sadly the outputs are disgusting on it, I can make an adapter but even that is far from ideal, might have to explore that a bit, but a sound card is my goal for a pre amp.
I've a cheap adapter made up with some 12 gauge monster cable that's over a dozen feet long and the audigy is able to drive it flawlessly though, no noise on it at all (audible) until I turn the computer off, then a 60hz hum takes over with a vengeance.
Regards,
Chris
Wytco0 ,
When you NAD has balanced output, that's great please use this!
The UcD is called a "digital amplifier" but there is nothing difgital inside....... It's an analog modulated PWM amplfier.
Regards,
Jan-Peter
www.hypex.nl
When you NAD has balanced output, that's great please use this!
The UcD is called a "digital amplifier" but there is nothing difgital inside....... It's an analog modulated PWM amplfier.
Regards,
Jan-Peter
www.hypex.nl
The shield is at least connected hard to the transmitting end. On the receiving end, it may be connected hard as well, or through a capacitor if the receiving equipment has a "pin 1 problem" i.e. if the shield is connected to the pcb instead of chassis.ewildgoose said:
For increased rejection of near-field M fields, star quad cables may be used. They are otherwise unpopular with pro people due to their higher capacitance.
Yes - in that case great emphasis must be put on the ability of differential receivers to reject common mode. And especially in that case, making sure that the shield current goes straight down the chassis instead of into the pcb is of paramount importance. A fully differential design can live with some unwanted earth current through the PCB, a single-ended circuit with balanced drivers/receivers can under no circumstances tolerate ground currents through the pcb. For that reason again: tie pin 1 to case.FuriousD said:
This is now an official AES recommendation. Neutrik is supplying XLR chassis parts with pin 1 firmly welded to the chassis.
A problem that can occur with the standard rca-to-xlr cable I described before is when a consumer device (floating gnd) is connected to two grounded balanced boxes. The equalising current flowing between the 2 grounded boxes goes through the consumer device. There, tiki's advice is sound: tie the shield to case, not to the rca ground. The better adapter cable is thus one where the shield comes out with a screw lug, and an rca plug with the hot wire to pin and cold to sleeve.
classd and Jan-peter, thanks for reminding me that these modules are not digital, silly me
My modules just arrived (arrived yesterday but I was out) they look very nice. Hoping to put them together later this week, I am now thinking of replacing the insides of my MF Typhoons with the UCD modules as I can do them one at a time. I might even be able to use the MF power supplies but I dont have any details of the schematic for them .
My modules just arrived (arrived yesterday but I was out) they look very nice. Hoping to put them together later this week, I am now thinking of replacing the insides of my MF Typhoons with the UCD modules as I can do them one at a time. I might even be able to use the MF power supplies but I dont have any details of the schematic for them .
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.