Hi JM,
At this time, I think build TA2050 in single rail..
You think it's better in dual in my case ?
Recharge it with one gel charger, batteries in parallel.
This system is not to listen to music all the day but for about ten hours per week.
Digital music : I use pc like a drive cause I'am working under ramdisk.
So I need a dac asynchronous usb with upsampling.
At this time, I think build TA2050 in single rail..
You think it's better in dual in my case ?
Recharge it with one gel charger, batteries in parallel.
This system is not to listen to music all the day but for about ten hours per week.
Digital music : I use pc like a drive cause I'am working under ramdisk.
So I need a dac asynchronous usb with upsampling.
I use "gel" batteries all the time (but not for audio). I was thinking messy as in having two or three separate batteries and the charging arrangements needed.I see "TA2050" has crept in above
Don't confuse me... do you mean TDA2050's or the Tripath TA2050 Class D amp.
Hi,
Below the schematic..
R3 is input impedance, I chose 27 Ko, but after if I want 10 or 47 just change value of R3..
Gain is 33, I read that it was 10 to 100, so 33 seems good for me..
C1 is Fc of input that mean about 3 Hz with 27 Ko and 2,2 uF..
Below the schematic..
R3 is input impedance, I chose 27 Ko, but after if I want 10 or 47 just change value of R3..
Gain is 33, I read that it was 10 to 100, so 33 seems good for me..
C1 is Fc of input that mean about 3 Hz with 27 Ko and 2,2 uF..
An externally hosted image should be here but it was not working when we last tested it.
That looks fine...
C7 could usefully be larger for extended bass response. I would suggest at least a 4700uf/35 volt cap and preferably 6800uf (particularly if the speaker impedance is low). Modern radial caps are small so there is no real problem. A 25 volt working voltage would be OK on 24 volt rails. The cap will see a steady no signal DC voltage of 24/2 (12 volts) and under signal conditions the amp output will stop a little bit short of the rails. C4 I would make at least 47uf.
To save keep working out cutoff frequencies try an on-line calculator,
calculators for electronic circuit design
C7 could usefully be larger for extended bass response. I would suggest at least a 4700uf/35 volt cap and preferably 6800uf (particularly if the speaker impedance is low). Modern radial caps are small so there is no real problem. A 25 volt working voltage would be OK on 24 volt rails. The cap will see a steady no signal DC voltage of 24/2 (12 volts) and under signal conditions the amp output will stop a little bit short of the rails. C4 I would make at least 47uf.
To save keep working out cutoff frequencies try an on-line calculator,
calculators for electronic circuit design
R6 and C6 are commonly known as a "Zobel" network.
Scroll down to the bottom of the page where it says "Zobel networks and loudspeaker drivers".
Zobel network - Wikipedia, the free encyclopedia
Almost all solid state amps have this network connected to the output and that brings a practical point up. The speaker ground return and the zobel return should really go back directly to the power supply main ground and not via the other components that also connect to ground. Thats because relatively high currents flow here and they can interfere and modulate the other points if they are all just connected "along the same piece of wire or PCB".
Values vary although I would stick to the recommended parts. They make no difference to the sonics of the amp and in practice its not that critical as long as they are there. 10 ohm and 0.1 uf are very common values for many discrete power amps. Without them many amps just oscillate wildy and destructively.
Scroll down to the bottom of the page where it says "Zobel networks and loudspeaker drivers".
Zobel network - Wikipedia, the free encyclopedia
Almost all solid state amps have this network connected to the output and that brings a practical point up. The speaker ground return and the zobel return should really go back directly to the power supply main ground and not via the other components that also connect to ground. Thats because relatively high currents flow here and they can interfere and modulate the other points if they are all just connected "along the same piece of wire or PCB".
Values vary although I would stick to the recommended parts. They make no difference to the sonics of the amp and in practice its not that critical as long as they are there. 10 ohm and 0.1 uf are very common values for many discrete power amps. Without them many amps just oscillate wildy and destructively.
The metal tab is connected to pin 3 and so is the ground connection in a single rail amp.
I'm going to say its very dependent on wiring and wiring practice how well both IC's would be with the common connection. There could be the possibility for interaction due to the small but finite resistances of the wiring.
Isolating kits are widely available. If they are mica you need thermal grease. If hey are synthetic material then often you don't. The package of the IC is a T0220 type.
Like this but there are many on offer...
MK3306/S - UNBRANDED - TRANSISTOR INSULATING KITS TO | CPC
I'm going to say its very dependent on wiring and wiring practice how well both IC's would be with the common connection. There could be the possibility for interaction due to the small but finite resistances of the wiring.
Isolating kits are widely available. If they are mica you need thermal grease. If hey are synthetic material then often you don't. The package of the IC is a T0220 type.
Like this but there are many on offer...
MK3306/S - UNBRANDED - TRANSISTOR INSULATING KITS TO | CPC
So in a nutshell: are we choosing the single supply option?
Are we still going biamplified?
OK, then run the TDA2050 from the +24V rail, the TA2020 plus the digital stuff from the +12V rail.
Agree on that?
Yes, in theory the lower 12V battery will last a little less than the upper one, no big deal.
You monitor voltage on the lower one, as soon as it falls to, say, 11.5V , you stop and recharge.
Maybe the upper one had an extra hour life. So what?
Anyway, since you will charge both in parallel (I like that ), they will automatically self equalize, the flatter one will take more current from the charger .
*If* you happened to charge them in series, yes, the upper one *might* be overcharged, the lower one undercharged, but that's not a problem now.
Are we still going biamplified?
OK, then run the TDA2050 from the +24V rail, the TA2020 plus the digital stuff from the +12V rail.
Agree on that?
Yes, in theory the lower 12V battery will last a little less than the upper one, no big deal.
You monitor voltage on the lower one, as soon as it falls to, say, 11.5V , you stop and recharge.
Maybe the upper one had an extra hour life. So what?
Anyway, since you will charge both in parallel (I like that
), they will automatically self equalize, the flatter one will take more current from the charger .*If* you happened to charge them in series, yes, the upper one *might* be overcharged, the lower one undercharged, but that's not a problem now.
Mooly,
I will use synthetic mica (no need thermal grease) and screw isolate :
I will use synthetic mica (no need thermal grease) and screw isolate :
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Hi All,
First, happy new year !
I received all componants, so works will be strarting..
I did not use pcb but a little piece of Altuglas and make connections on "air" with gold plated connectors.
First, happy new year !
I received all componants, so works will be strarting..
I did not use pcb but a little piece of Altuglas and make connections on "air" with gold plated connectors.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.