I expect that someone who ist designing a amplifier,
also have an osciloscope to measure the most important Signals and correct If Something ist Not correct.
Or chose the right values
But most of Them are 1:1 copys of a cheap working amplifier ^^
..a Bit of a fast living , buy and replace "world"...
Gnarf
also have an osciloscope to measure the most important Signals and correct If Something ist Not correct.
Or chose the right values
But most of Them are 1:1 copys of a cheap working amplifier ^^
..a Bit of a fast living , buy and replace "world"...
Gnarf
Is not so bad. There are JLE, there are 3e
Products | jlelectronicsph
Do you mean that in LM2575 they give bad LC filters on the output or they do not give them at all?
Last edited:
TPA3255 board
Hello
I would like to inquire about this panel, what quality does it represent?
https://i.imgur.com/CNLoVxy.jpg
Thx
Hello
I would like to inquire about this panel, what quality does it represent?
https://i.imgur.com/CNLoVxy.jpg
Thx
Aiyima A-08 A08 built in power supply
The power supply built into the Aiyima A08 is probably adequate but not great.
The problem with this switching supply comes from the switching noise. I bought same AD-35036-52 supply to use with my Aiyima A07. On test, I found the switching frequency of the supply was 66kHz, each cycle produced a 'ringing' pulse of 20MHz that decayed within 500nS. More details can be found in my previous post #4189, I included oscilloscope traces there.
So the power supply is not great but again I think it is great value and perhaps in-line with the cost of the amplifier. So far I am very pleased with the price vs performance.
My attempts to reduce the switching spikes had only a very small impact. The problem with any add on filter is the balance between a high impedance (attenuation) for the pulses and a low impedance path (low loss) for the DC supply. On the PSU output I fitted an RC filter and ferrite choke.
I also found on the A07 that the TPA3255 switching noise was poorly contained and I suspect that the filtering between the analogue and digital sections are implemented badly. I hope that the A08 is better in this regard. A photo of power supply with my added filtering is attached.
The power supply built into the Aiyima A08 is probably adequate but not great.
The problem with this switching supply comes from the switching noise. I bought same AD-35036-52 supply to use with my Aiyima A07. On test, I found the switching frequency of the supply was 66kHz, each cycle produced a 'ringing' pulse of 20MHz that decayed within 500nS. More details can be found in my previous post #4189, I included oscilloscope traces there.
So the power supply is not great but again I think it is great value and perhaps in-line with the cost of the amplifier. So far I am very pleased with the price vs performance.
My attempts to reduce the switching spikes had only a very small impact. The problem with any add on filter is the balance between a high impedance (attenuation) for the pulses and a low impedance path (low loss) for the DC supply. On the PSU output I fitted an RC filter and ferrite choke.
I also found on the A07 that the TPA3255 switching noise was poorly contained and I suspect that the filtering between the analogue and digital sections are implemented badly. I hope that the A08 is better in this regard. A photo of power supply with my added filtering is attached.
Attachments
guys!!! if I can advise you, forget about any switching power supply ... make a good linear with transformer + diode bridge + capacitors and the TPA3255 will blossom like a rose! I have the Shui Yuan board with the LME49720NA and I always used it with a 36V 5A switching power supply, and nothing to say...was fine ... but then I put on a power supply with a toroidal and it's all music, much more airy, soft, the soundstage is deeper, wider, more real!!
yes...I know there is switching and switching .... but if I have to buy expensive switching, a linear power supply is better
this is mine
1PCS Convertitore AC 110V 220V DC 36V 5A 180W Alimentatore Stabilizzato Trasformatore di Alimentazione del Driver DC36V 5 di Commutazione di Alimentazione|Switching Power Supply| - AliExpress
this is mine
1PCS Convertitore AC 110V 220V DC 36V 5A 180W Alimentatore Stabilizzato Trasformatore di Alimentazione del Driver DC36V 5 di Commutazione di Alimentazione|Switching Power Supply| - AliExpress
You call a 15U$ Dollar smps from Ali expensive? oh well
no, I have not explained myself ... this SMPS is a power supply with a cost aligned with that of the card, it is as cheap as the card ... it is not worth spending for an expensive SMPS, from 50 or 100 euros, because it is much better spend less for a good linear supply, that is what I mean
Regarding switching power supplies, cheap and expensive. There is a genuine difference in performance, not simply the cost.
Cheap supplies use a simple square wave switching technology resulting in unwanted spikes at the output. A more complex design still has the 'chopped' (switching) circuits, but the resultant energy is fed into resonant circuit. The output of the high energy resonant circuit are now sine waves, these are easily scaled to the desired output voltage before rectification to DC. The output does not suffer from the spikes seen in the non-resonant supply. The newer SMPS should be described as 'resonant', a typical example is an LLC resonant supply.
Linear is very good but the weight might be a factor. A big transformer with a lot of caps might need an inrush current limiter. Yes a linear is good, but SMPS are on the rise and there are good, but costly items available.
Cheap supplies use a simple square wave switching technology resulting in unwanted spikes at the output. A more complex design still has the 'chopped' (switching) circuits, but the resultant energy is fed into resonant circuit. The output of the high energy resonant circuit are now sine waves, these are easily scaled to the desired output voltage before rectification to DC. The output does not suffer from the spikes seen in the non-resonant supply. The newer SMPS should be described as 'resonant', a typical example is an LLC resonant supply.
Linear is very good but the weight might be a factor. A big transformer with a lot of caps might need an inrush current limiter. Yes a linear is good, but SMPS are on the rise and there are good, but costly items available.
The tradeoff is efficiency. A linear psu at 36A 5A is 180 watts, if you have a max of 60% efficiency then something like 72 watts are wasted by the linear PSU's low efficiency. A 200W Meanwell LRS-200-36 is about 90% efficient, so you are only wasting 20 watts.
I'm surprised that some Chinese outfit hasn't put out a SMPS ripple reducer for $15 that you can put in between a SMPS and your stereo.
I am using Aiyima A07 , a tpa 3255 based amp, with black brick adapter 32v 5A , then I bought 48v 5A for more watt. Recommended psu for this amp is 48v 7.5A.
I am interested in buying Meanwell LRS-350-48 , but I think 48v is kinda hot for this amp. The casing temperature is 36.8°C using 32v 5A , volume at 9 o'clock, but when using 48v 5A, the casing temp increases to 38.9°C volume at 9 o'clock, and increases to 40°C with volume at 10-11 o'clock.
Meanwell LRS 350-48 can be set to 36v with 9.7A, my question is, can this amp use 9.7A ?? recommended is 48v and 7.5A. I think with lower voltage , the amp will be less hot. The position of the Electrolytic capacitor in this amp is really close to the heatsink and it's only 50v so with 36v that would leave more room before max, let the adapter work harder than the amplifier.
Another question that I'd like to ask is the Electrolytic capacitor in this amp, it is using only 2x1000µF = 2000µF, is it worth it to use adapter with more power ? Most high watt amplifiers usually have really high Electrolytic capacitor. A vid on youtube says it should be 2200µF per 1A, but he's talking about classAB. Is that true ? if it is does that only apply to class ab or general ?
One more thing (sorry for double posting, please merge, I'm new to the forum and still moderated) is it safe to leave Aiyima A07 turned ON with 32v5A or 48v5A black brick adapter for 16H a day every day ?
(for comparison I usually left my Klipsch Promedia 2.1 ON for 24/7/365 , it's been years like that, but not using it now, it's still working perfectly though. Another comparison with black brick adapters are modem and adsl router, they're on 24/7/365 with no prob for years)
I am interested in buying Meanwell LRS-350-48 , but I think 48v is kinda hot for this amp. The casing temperature is 36.8°C using 32v 5A , volume at 9 o'clock, but when using 48v 5A, the casing temp increases to 38.9°C volume at 9 o'clock, and increases to 40°C with volume at 10-11 o'clock.
Meanwell LRS 350-48 can be set to 36v with 9.7A, my question is, can this amp use 9.7A ?? recommended is 48v and 7.5A. I think with lower voltage , the amp will be less hot. The position of the Electrolytic capacitor in this amp is really close to the heatsink and it's only 50v so with 36v that would leave more room before max, let the adapter work harder than the amplifier.
Another question that I'd like to ask is the Electrolytic capacitor in this amp, it is using only 2x1000µF = 2000µF, is it worth it to use adapter with more power ? Most high watt amplifiers usually have really high Electrolytic capacitor. A vid on youtube says it should be 2200µF per 1A, but he's talking about classAB. Is that true ? if it is does that only apply to class ab or general ?
One more thing (sorry for double posting, please merge, I'm new to the forum and still moderated) is it safe to leave Aiyima A07 turned ON with 32v5A or 48v5A black brick adapter for 16H a day every day ?
(for comparison I usually left my Klipsch Promedia 2.1 ON for 24/7/365 , it's been years like that, but not using it now, it's still working perfectly though. Another comparison with black brick adapters are modem and adsl router, they're on 24/7/365 with no prob for years)
Last edited:
Corrections here...
A big transformer, rectifier and big caps used to drive a class D amp is going to be more than 95% efficient. The only loss here is due to the rectifier. At an average of 40W per channel the PSU efficiency is going to be about 99%.
SMPS switching noise is high frequency ringing and not easy to get rid of. Ripple on the other hand is a low frequency drop in the supply voltage between output pulses. Ripple is relatively easy to get rid of but that is not normally a problem with SMPS.
A big transformer, rectifier and big caps used to drive a class D amp is going to be more than 95% efficient. The only loss here is due to the rectifier. At an average of 40W per channel the PSU efficiency is going to be about 99%.
SMPS switching noise is high frequency ringing and not easy to get rid of. Ripple on the other hand is a low frequency drop in the supply voltage between output pulses. Ripple is relatively easy to get rid of but that is not normally a problem with SMPS.