UcD400 Q & A

Jump to Latest
#882
dead time

Hi Bruno.

Thanks for the swift reply, I always thought that the faster the switching capability of the MosFets the better.

Do you agree that it doesn't really matter what kind of output stage is used (SE,bridged etc) since the powerstage works basically in class B or zero bias and the distortion introduced by it is minimal?

I read you are working on a low power "high end" version of the UcD modules, any projected specs yet?

regards.

Kees.
 
#884
Re: dead time

kro5998 said:
Do you agree that it doesn't really matter what kind of output stage is used (SE,bridged etc) since the powerstage works basically in class B or zero bias and the distortion introduced by it is minimal?

I read you are working on a low power "high end" version of the UcD modules, any projected specs yet?
Click to expand...
The reasons for choosing between half or full bridge are practical in nature, in principle both are equally good. It is often a good choice for very high power levels but the need for good resistor matching in the feedback path of full-bridge amplifiers is a pain. The current Hypex modules are single-ended because we know some users will want to bridge them for higher power so we've got more potential applications covered with less products.

The "high end" thing is just a muscle flexing exercise for low distortion amps, not necessarily low power. I'm targeting 0.003% / all frequencies.
 
#887
Re: Re: dead time

Bruno Putzeys said:

The reasons for choosing between half or full bridge are practical in nature, in principle both are equally good. It is often a good choice for very high power levels but the need for good resistor matching in the feedback path of full-bridge amplifiers is a pain. The current Hypex modules are single-ended because we know some users will want to bridge them for higher power so we've got more potential applications covered with less products.

The "high end" thing is just a muscle flexing exercise for low distortion amps, not necessarily low power. I'm targeting 0.003% / all frequencies.
Click to expand...

Hi Bruno,

Wouldn't a fully differential amp like the Hypex half bridge versions of UCD benefit from tight tolerance components in the feedback loops as well, or is it just less critical in the half bridge?

Thanks,
Chris
 
#888
Re: Re: Re: dead time

classd4sure said:
Wouldn't a fully differential amp like the Hypex half bridge versions of UCD benefit from tight tolerance components in the feedback loops as well, or is it just less critical in the half bridge?
Click to expand...
The half bridge is by all means less sensitive to component tolerances. Component mismatch leads only to a CMRR degradation. On a full bridge version component mismatch will result in degradation of PSRR and THD as well. Ouch!
 
#889
Hi, I am currently building the PSU for the UCD400 and I am aware that the max voltage that the module can take is 63V. I remember somewhere in the forum that VDR can be used to limit the occasional power surge.

Can someone please direct me which part of the circuit should I place the VDR and which model of the VDR should I use?
 
#894
#897
chrisb03 said:



Isn't there a problem using this circuit as it uses common ground?

Here's a few other options which I think have been knocked back due to common ground.

http://www.ampslab.com/dcpro.htm

http://www.wnaudio.com/lspro.html

Isn't there a straight out answer for implementing speaker protection? Bruno, JP?
Click to expand...

When you make a DC protection circuit it is not critical where you ground it. After all, the only thing this circuit is interested in, is whether a DC voltage of several volts exists. No grounding scheme can be so bad it interferes with that function. It's easiest to ground the DC protection circuit at the power supply.

From the output line of each amp, take a 47k resistor to a common 100uF capacitor (a summing lowpass filter so to speak). Make a circuit (takes 3 transistors) to trigger when the voltage across this cap becomes + or - Vbe. This will not trigger on a 10Hz full scale sine wave, but it will go off within 50ms when the amp goes bad.
Again, the ground of this circuit is taken from the power supply ground.

Make sure the relay stays off when the protection triggers (until the plug is pulled).
 
#898
Bruno Putzeys said:


When you make a DC protection circuit it is not critical where you ground it. After all, the only thing this circuit is interested in, is whether a DC voltage of several volts exists. No grounding scheme can be so bad it interferes with that function. It's easiest to ground the DC protection circuit at the power supply.

From the output line of each amp, take a 47k resistor to a common 100uF capacitor (a summing lowpass filter so to speak). Make a circuit (takes 3 transistors) to trigger when the voltage across this cap becomes + or - Vbe. This will not trigger on a 10Hz full scale sine wave, but it will go off within 50ms when the amp goes bad.
Again, the ground of this circuit is taken from the power supply ground.

Make sure the relay stays off when the protection triggers (until the plug is pulled).
Click to expand...



Thanks Bruno. My electronics isn't that good, I think I have understood what your saying, but I prefer to buy something ready made as I don't have the time. Is this circuit ok http://www.ampslab.com/dcpro.htm? If so, do I have to have a separate psu? (I have dual monitor)

Thanks again
 
You must log in or register to reply here.
Top Bottom