do the lights go dim when you turn it on? 
Iv'e been using these:
http://www.ska-audio.com/diy/softstart.html
The kit is very cheap and effective. No need for an additional winding or a separate transformer.
Maybe christi could come up with a connexelectronics softstart
col.
Iv'e been using these:
http://www.ska-audio.com/diy/softstart.html
The kit is very cheap and effective. No need for an additional winding or a separate transformer.
Maybe christi could come up with a connexelectronics softstart
col.
Keep it simple with 1 passive component, it is called NTC.
See link ; http://www.ametherm.com/Inrush_Current/inrush_current_faq.html
These are used by many manufacturer and they work great.
Eric
See link ; http://www.ametherm.com/Inrush_Current/inrush_current_faq.html
These are used by many manufacturer and they work great.
Eric
finished my TA3020
Finally managed to get rid of the ground loop/hum and now have my TA3020 in my tri-amp system powering my compression drivers. Some photos of the amp are here:
http://minirig.org.au/2009/02/14/minirig-connexelectronics-tripath-ta3020/
please leave comments on the blog
col.
Finally managed to get rid of the ground loop/hum and now have my TA3020 in my tri-amp system powering my compression drivers. Some photos of the amp are here:
http://minirig.org.au/2009/02/14/minirig-connexelectronics-tripath-ta3020/
please leave comments on the blog
col.
No, I have no overheating problems at all, the heatsink is warm but not hot, I can easily leave my hand on it without having to take it off.
I am only running my amp at low voltages (42-43v) though and at the moment it is only powering the high frequency drivers in a tri-amp system, so it really is not doing a lot of work. Also, my case has very good ventilation.
col.
(Old version) TA2022 board connections
Hi Cristi,
Now that I have finished my TA3020 amp I am moving onto my TA2022 board. I have another dumb question you have answered it before but I'm still not sure. The PCB is very small and double sided plate-through so I can't quite tell where the tracks are going.
Which side is left and which side is right, If I am looking at the side with the components mounted and the Tripath chip on the top?
cheers,
col.
P.S. any estimated time of arrival for the DA-1 SMPS?
Hi Cristi,
Now that I have finished my TA3020 amp I am moving onto my TA2022 board. I have another dumb question
you have answered it before but I'm still not sure. The PCB is very small and double sided plate-through so I can't quite tell where the tracks are going.Which side is left and which side is right, If I am looking at the side with the components mounted and the Tripath chip on the top?
cheers,
col.
P.S. any estimated time of arrival for the DA-1 SMPS?
Hi Col,
as you look from above, to the components side with the IC in front, and the connectors the left side is left, and right side is yes, right. you need to take care about the GND connection, since there are 2 GND, one for signal, one for power. depending on their junction point, is also the SN ratio.
as you look from above, to the components side with the IC in front, and the connectors the left side is left, and right side is yes, right. you need to take care about the GND connection, since there are 2 GND, one for signal, one for power. depending on their junction point, is also the SN ratio.
This IC is internally limited to 8A so at +- 20Vac, 10A (400VA) you are OK.
I'm presently building the same setup with a 400VA (also from Antek) but using 1 x TDA8920BTH per output in bridge mode (6 ohm min.) I will be putting a smoked fiberglass on top so I can see the gts of the amp. Should be nice.
Eric
I'm presently building the same setup with a 400VA (also from Antek) but using 1 x TDA8920BTH per output in bridge mode (6 ohm min.) I will be putting a smoked fiberglass on top so I can see the gts of the amp. Should be nice.
Eric
Hi,
I brought my amplifier to work to get it tested, the amp used is 2 x TDA8920BTH sold by Connexelectric.
With both inputs shorted I have 5.5Vrms at the output, the frequency is 348kHz (not a typo). Looks like the output filter is not designed properly or maybe not made for bridge into 8 ohm.
Even when trying to get 10 or 100W at 8 ohm at the output the 5Vrms high carrier frequency is still there. I was not able to get a nice signal since the 347kHz is always there. Could this be a bad batch of PCB ? My heatsink is grounded as per the recommendation of the Philips datasheet.
You do not hear it of course (maybe dogs do...) but now knowing that it is there will drive me crazy.
Is it possible to get the schematic so I can send it to my friend who is a Class D amp designer.
Connexelectric, please let me know what you can do to fix this.
This amp is not supplied with an input connector therefore I have soldered the input directly.
On a positive note : The amps sounds good.
Thanks and best regards,
Eric
I brought my amplifier to work to get it tested, the amp used is 2 x TDA8920BTH sold by Connexelectric.
With both inputs shorted I have 5.5Vrms at the output, the frequency is 348kHz (not a typo). Looks like the output filter is not designed properly or maybe not made for bridge into 8 ohm.
Even when trying to get 10 or 100W at 8 ohm at the output the 5Vrms high carrier frequency is still there. I was not able to get a nice signal since the 347kHz is always there. Could this be a bad batch of PCB ? My heatsink is grounded as per the recommendation of the Philips datasheet.
You do not hear it of course (maybe dogs do...) but now knowing that it is there will drive me crazy.
Is it possible to get the schematic so I can send it to my friend who is a Class D amp designer.
Connexelectric, please let me know what you can do to fix this.
This amp is not supplied with an input connector therefore I have soldered the input directly.
On a positive note : The amps sounds good.
Thanks and best regards,
Eric
Eric
all class D or class T amplifier have a residual component with same frequency as the carrier at output. this cannot be completely eliminated, but can be attenuated further more by using higher order output filters. since this component is far away form the hearing spectrum, cannot be hear at all. not even dogs can hear that high. think about that there are some low power filterless class D amplifiers on the market, which have the amplitude of that component equal to the full voltage swing, going into the loudspeakers, 5-28V, not 5 mV.
also, consider that if you made the measurement without load, you may experience some resonance in the output filter which gave you that value. try to connect in paralel woth the output a 4 ohm resistor, or the loudspeaker.
a first order filter which can attenuate from 50Vvv to 5 mV (calculate how many dB) i think is calculated right. even in some class D amplifiers IC datasheets they mention about this residual component.
the pcb is designed carefully, avoiding GND loops and ringing.
schematic can be found in the manual, which you can download from here: http://www.connexelectronic.com/documents/TDA892x_Audio_Amplifier_Module.pdf
all class D or class T amplifier have a residual component with same frequency as the carrier at output. this cannot be completely eliminated, but can be attenuated further more by using higher order output filters. since this component is far away form the hearing spectrum, cannot be hear at all. not even dogs can hear that high. think about that there are some low power filterless class D amplifiers on the market, which have the amplitude of that component equal to the full voltage swing, going into the loudspeakers, 5-28V, not 5 mV.
also, consider that if you made the measurement without load, you may experience some resonance in the output filter which gave you that value. try to connect in paralel woth the output a 4 ohm resistor, or the loudspeaker.
a first order filter which can attenuate from 50Vvv to 5 mV (calculate how many dB) i think is calculated right. even in some class D amplifiers IC datasheets they mention about this residual component.
the pcb is designed carefully, avoiding GND loops and ringing.
schematic can be found in the manual, which you can download from here: http://www.connexelectronic.com/documents/TDA892x_Audio_Amplifier_Module.pdf
Hi CNX,
Thanks for the quick reply.
The residual carrier frequency is 5.5Vrms not 5mV, if it was 5mV I would not care.
I tested the amp with an 8 ohm per output since this is a bridge design using one module per output.
Is it possible to have the schematic without the watermark, then I could send the schematic to my friend and he will let me know which modification he suggest to lower this residual carrier voltage.
p.s. You are correct about the ground loop, this amp is extremly quiet and I can barely hear the noise floor with my 94db speaker. Great work on the layout.
Thank you,
Eric
Thanks for the quick reply.
The residual carrier frequency is 5.5Vrms not 5mV, if it was 5mV I would not care.
I tested the amp with an 8 ohm per output since this is a bridge design using one module per output.
Is it possible to have the schematic without the watermark, then I could send the schematic to my friend and he will let me know which modification he suggest to lower this residual carrier voltage.
p.s. You are correct about the ground loop, this amp is extremly quiet and I can barely hear the noise floor with my 94db speaker. Great work on the layout.
Thank you,
Eric
- Status
- Not open for further replies.