Does anyone else get concerned about seeing LC filters moved off-board on amps such as from Sure Electronics?
I just checked out a AA-AS32171 module from Sure that uses TDA7492, and looked through a few threads on mods to pcbs with that and similar ST chips. I can appreciate the issues with parts selection for the output LC filtering and distortion, but when dealing with squarewaves and a few hundred kHz fundamental at many amps, its a big ask to move the position of such parts off-board.
If an aim is to use a better output filter cap for distortion reduction, and there is no opportunity to place the new cap directly to the pcb pads, then it may be wise to fit a smaller smt ceramic on the pcb pads to still give the shortest path for higher frequencies.
I just checked out a AA-AS32171 module from Sure that uses TDA7492, and looked through a few threads on mods to pcbs with that and similar ST chips. I can appreciate the issues with parts selection for the output LC filtering and distortion, but when dealing with squarewaves and a few hundred kHz fundamental at many amps, its a big ask to move the position of such parts off-board.
If an aim is to use a better output filter cap for distortion reduction, and there is no opportunity to place the new cap directly to the pcb pads, then it may be wise to fit a smaller smt ceramic on the pcb pads to still give the shortest path for higher frequencies.
A diy LISN may be one way to throw some measureable 'hi-fi performance' at the out-of-band frequency response of the amp. I mean many on these forum boards are quite concerned about operating frequencies extending up to 100kHz - why not extend that to 1Mhz. At the levels of emi likely to be generated from some of the amp photos I've seen, you wouldn't need a $10k HP spectrum analyser to pick out the fundamental and harmonics - just use a left-over SiC diode from the power supply to rectify the signal and measure the DC volts 
Or resurrect an old handheld AM radio with a tuning meter and watch the meter needle get pinned to the end stop.
Or resurrect an old handheld AM radio with a tuning meter and watch the meter needle get pinned to the end stop.
Speaking as a ham radio operator, broadcast equipment designer and someone who has pushed multiple products through CE... please don't do this unless you absolutely know what you're doing.
You want the power stage ground -> power stage switching node -> inductor -> filter capacitor -> filter capacitor ground -> power stage ground loop to be as short and tight as possible, to minimize any possible radiation.
If you build this, do the following:
- Remove the filter inductor/capacitor on the amplifier.
- Jumper over the inductor location on the PCB, or leave the inductor there.
- Make some tight (several turns/inch) twisted pair wire to go between your external filter and the amp card. Connect this to the location on the board where the filter capacitor was.
- Run this through a common mode choke before you bring it to your external filter. Make sure the choke has lots of impedance at low frequencies (10-100KHz), don't just use one that you don't know anything about. This choke won't affect audio performance.
- Make sure the new output filter is very tightly constructed.
- Put everything in a metal case.
But before you do this, make a good justification for what you're doing. 9 chances out of 10 you'll be able to bodge a new inductor/capacitor onto the existing PCB.
You want the power stage ground -> power stage switching node -> inductor -> filter capacitor -> filter capacitor ground -> power stage ground loop to be as short and tight as possible, to minimize any possible radiation.
If you build this, do the following:
- Remove the filter inductor/capacitor on the amplifier.
- Jumper over the inductor location on the PCB, or leave the inductor there.
- Make some tight (several turns/inch) twisted pair wire to go between your external filter and the amp card. Connect this to the location on the board where the filter capacitor was.
- Run this through a common mode choke before you bring it to your external filter. Make sure the choke has lots of impedance at low frequencies (10-100KHz), don't just use one that you don't know anything about. This choke won't affect audio performance.
- Make sure the new output filter is very tightly constructed.
- Put everything in a metal case.
But before you do this, make a good justification for what you're doing. 9 chances out of 10 you'll be able to bodge a new inductor/capacitor onto the existing PCB.
I sold a class d amp on ebay.
The buyer rang up and told me it didn't work.
We eventually tracked it down to him removing mosfets off the pcb and putting them on an external heatsink.
Of course putting them back on the pcb fixed his problems.
A few extra inches of wire caused havoc. While the square wave is a couple of kilohertz that is only the fundamental frequency and the wave consists of many harmonics.
The buyer rang up and told me it didn't work.
We eventually tracked it down to him removing mosfets off the pcb and putting them on an external heatsink.
Of course putting them back on the pcb fixed his problems.
A few extra inches of wire caused havoc. While the square wave is a couple of kilohertz that is only the fundamental frequency and the wave consists of many harmonics.
Interesting, I've been looking for threads on this.
I just bought the 2x25 watts version off parts express. Listening to it burn in with a 15 volt 6 amp switching power supply for 6 hours yesterday through a pair of 8 ohm 4" full range speakers I'm impressed.
Some early notes;
With a 12 volt 2 amp power supply this amp is useable expect distortion to start early if you attempt to get loud.
I bought this looking for a low energy power amp and this will work for that.
I ordered a 19 volt 4 amp laptop supply, supposedly optimal according to most threads I have seen.
Presently the board is getting power through a 15.2 volt 6 amp power supply.
I did notice the directions are next to impossible to figure out since the board I have is different then the one the directions were printed for.
I also ordered the volume encoder board, I plan on installing that today.
I kept adjusting the 2 gain dip switches and settled on low. (weak, low medium and high as options). I really could not hear any increase in volume adjusting the dip switched so I went for best bass response.
I am using a sony walkman CD player as input. Started out with the volume at just 3 bars, 1/2 way through the CD I turned the sony up to 6 with 3 bars and that is as loud as I wanted to listen to it. Still clear but plenty loud enough.
After a CD I switched to my MP3 player (no mp3's under 320) and pushed the volume up till I was pushing a steady average of 1.5 volts and the I Pad sound analyzer was showing 90-95db depending on the material. It stayed there for 5 hours while I played around in the garage. Heat sink was never warm so I doubt I was pushing this too hard.
I did sit down and listen carefully at several times during the evening.
So far I am very impressed.
They have a 2x50 watt version, looks the same but with a fan, from reading the datasheet it looks like all you get with that extra volume is extra distortion. There is also a 100 watt version. Naturally you need bigger power supplies.
I just bought the 2x25 watts version off parts express. Listening to it burn in with a 15 volt 6 amp switching power supply for 6 hours yesterday through a pair of 8 ohm 4" full range speakers I'm impressed.
Some early notes;
With a 12 volt 2 amp power supply this amp is useable expect distortion to start early if you attempt to get loud.
I bought this looking for a low energy power amp and this will work for that.
I ordered a 19 volt 4 amp laptop supply, supposedly optimal according to most threads I have seen.
Presently the board is getting power through a 15.2 volt 6 amp power supply.
I did notice the directions are next to impossible to figure out since the board I have is different then the one the directions were printed for.
I also ordered the volume encoder board, I plan on installing that today.
I kept adjusting the 2 gain dip switches and settled on low. (weak, low medium and high as options). I really could not hear any increase in volume adjusting the dip switched so I went for best bass response.
I am using a sony walkman CD player as input. Started out with the volume at just 3 bars, 1/2 way through the CD I turned the sony up to 6 with 3 bars and that is as loud as I wanted to listen to it. Still clear but plenty loud enough.
After a CD I switched to my MP3 player (no mp3's under 320) and pushed the volume up till I was pushing a steady average of 1.5 volts and the I Pad sound analyzer was showing 90-95db depending on the material. It stayed there for 5 hours while I played around in the garage. Heat sink was never warm so I doubt I was pushing this too hard.
I did sit down and listen carefully at several times during the evening.
So far I am very impressed.
They have a 2x50 watt version, looks the same but with a fan, from reading the datasheet it looks like all you get with that extra volume is extra distortion. There is also a 100 watt version. Naturally you need bigger power supplies.
It's worth checking in these amps for heatsinking design. I added two metal shims with heatsink interface paste to just wedge the pcb TDA7492 module underside heatsink to the chassis aluminium floor - otherwise there was a 2-3mm gap.
It's a nice module, with supply input via series diode and CM choke, and then LC filter (C=4x 220uF 50V 105C plus smt).
The only other change I thought worth making was to squeeze in a zobel on each speaker output to maintain some hf loading, and aid in constraining switching noise to the amp itself.
It's a nice module, with supply input via series diode and CM choke, and then LC filter (C=4x 220uF 50V 105C plus smt).
The only other change I thought worth making was to squeeze in a zobel on each speaker output to maintain some hf loading, and aid in constraining switching noise to the amp itself.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.