I've a plastic chassis (3D printed 😛)... so no ground. Could be that?
Furthermore, I use the ampli with my PC. So it happens that, whilst the ampli is always on, the PC and USB card go in stand-by. So, when I wake up the PC, the pops arise. Is there any way to prevent it? Could a "speaker protection board" be of any help?
Another question: is it only me?! 😕😕😕
The pc pops, so everybody that connects a similar pc to any poweramplifier that is on, will hear what you hear. (Related thinking out of the box, I improved dieselconsumption and emissions of my VW TDI by taking bikes, the bikes save fuel and have zero emission. You do hear the wind a little on highway, but that is only slight offer when I do so much good for environment. Bikes were unused standing in garage, so cleans up space in garage too, a real win win)
Modify
Is it anything to mod with this board?
Very dissapointed about the harschness.
Bootsnubber maybe?
But where and what values?
Is it anything to mod with this board?
Very dissapointed about the harschness.
Bootsnubber maybe?
But where and what values?
My suspicions would fall on the opamps being infected with switching noise as the most likely cause of harshness. What are the opamps doing, circuit-wise ? Are they tone controls?
<edit> OK found the schematic on the previous page 🙂
<edit> OK found the schematic on the previous page 🙂
Last edited:
Yeah, me too!
Me too Clengman. I'd like to know what is so horrible about these two boards. Because both the boards sound fine to me!
The red dual chip 3116 board I own had some hiss. So, I always used it with the Guanzo two tube preamp with the volume pot on the amp cranked to maximum. And used the volume pot on the preamp to control the level.
But a friend of mine said to remove dual chip's volume pot and on/off. I replaced the stock volume pot with a 10K stereo one. (Original was 50K which he thinks is a linear, and not an audio pot).
Now the dual chip amp is whisper quiet! And I can use it "stand alone" with no preamp.
As for the blue/black board. Originally I didn't like it much, but gave it a second listen once I got the Guanzo (which people dismiss as well but to me sounds just fine). I think originally I was running the Blue Black with just a straight passive preamp and that just wasn't a right match for it.
There is a page someplace here which has photos of all the amps and what they are known by: blue/black, YJ 2.0, etc. That is helpful.
Mark
Me too Clengman. I'd like to know what is so horrible about these two boards. Because both the boards sound fine to me!
The red dual chip 3116 board I own had some hiss. So, I always used it with the Guanzo two tube preamp with the volume pot on the amp cranked to maximum. And used the volume pot on the preamp to control the level.
But a friend of mine said to remove dual chip's volume pot and on/off. I replaced the stock volume pot with a 10K stereo one. (Original was 50K which he thinks is a linear, and not an audio pot).
Now the dual chip amp is whisper quiet! And I can use it "stand alone" with no preamp.
As for the blue/black board. Originally I didn't like it much, but gave it a second listen once I got the Guanzo (which people dismiss as well but to me sounds just fine). I think originally I was running the Blue Black with just a straight passive preamp and that just wasn't a right match for it.
There is a page someplace here which has photos of all the amps and what they are known by: blue/black, YJ 2.0, etc. That is helpful.
Mark
Its NE5532P opamps, don't have anything other to switch to.But you they are so bad?
Opamps aren't bad in and of themselves, just they're a bit tricky to get sounding really good.
In this case I see that there's a master volume control which has been given rather a heavy load caused by the opamp circuit beyond it being inverting. Not sure why they chose that configuration but its not a good one for sound quality. Those 4k7 resistors connected to opamp A really need to be swapped out for 47ks (at the minimum).
In this case I see that there's a master volume control which has been given rather a heavy load caused by the opamp circuit beyond it being inverting. Not sure why they chose that configuration but its not a good one for sound quality. Those 4k7 resistors connected to opamp A really need to be swapped out for 47ks (at the minimum).
Your photo and the schematic don't seem to agree - you're showing 33k resistors but there are none of that value in the schematic. So I'm lost....
The power supplies to the opamps would be the next area to attack but they're not shown on the schematic either.
The power supplies to the opamps would be the next area to attack but they're not shown on the schematic either.
Are you having troubles regarding 'pop' or noise/hum from the boards ?
IMHO PC doesn't cause any interference. Like what I did, I run an output direct from my PC to the TPA and it just dead quiet with my 83dB speaker.
Maybe a photo description of the board wiring inside the case would make clear picture.
Thanks!
It´s getting better!
Changed 4 resitors to 47k.
I don´t know about these?
This board is very similar to mine albeit with different values. To the right of the op-amp on the right ... The 4K7 and 33K resistors in conjunction with the 1uf decoupling caps (on the input after the volume control) and the 100pf caps on the underside of the board implement an inverting audio frequency bandpass filter at 34hz and 48Khz. Unfortunately this filter is only on the ".1" portion of the board.
The two 4K7 resistors you show replaced with 47K are resistors is the ".1" channels LP filter. That Fc of that filter frequency has now be reduced 10x from the previous Fc. I can't read the two black cap values that determine the filter's Fc with the original 4K7 resistors (my board used 10K resistors and .22uf/.1uf to implement a 107hz 2nd order Butterworth LP filter). You don't show the other two 4K7 resistors that you indicated you swapped for 47K. If they are to the left of the left op-amp, they are simply buffer resistors between the final op-amp stages and the TPA3116 input. I believe my board is running the TPA3116 at 26db gain. That would specify the input impedance to the TPA 3116 at 30K ohms (TI spec). So you may have reduced the signal going into the chip with the higher resistor value.
If you replaced the 4K7 resistors next to the 33K resistors on the right with 47K, you've changed the lower Fc of the ".1" channel's audio bandpass filter from 34hz to 3.4hz (that would correct the problem others have had on the missing low end frequencies on the sub channel). I use that filter "as-is" to mitigate X-max issues on my sub at high power).
You say it sounds better ... but if my analysis is correct, you've only modified the ".1" channel and I presume the harshness you mention is on the L-R channels as the ".1" channel doesn't produce any high frequencies. Having said (all) that, perhaps abraxalito's comment on the 4K7 resistors (rightmost ones) has an impact on the overall sound that I do not understand ... their effect on the op-amp circuit nor how it would effect the L-R channels.
Lastly, my board is dead quiet. No turn-on, turn-off pops whether using the boards switch or just plugging in the PS to the DC jack. I have no pop issues starting/pausing/restarting media player on my PC (nor playing mp3s on my phone). Never used it having the PC come out of hibernate so that may be the issue.
I am keenly interested on abraxalito's take on my post.
Last edited:
My boards subchannel I think is just that, no x-over what I can hear, only can control volume with the middle knob.Seems to be variants of this board with x-over.
Maybe just placebo my change of resistors but somehow it feels a little better in the harchness, not as good as my tripath but better.
Yes, the middle knob is only a volume control. It does not alter the sub's x-over frequency. The sub's x-over is fixed by the values of two resistors and two capacitors. What are values of the two black capacitors between the op-amps (can't read there markings in the pic).
Do you have a driver hooked up to the sub channel? If so, is it producing frequencies high frequency audio (>1Khz)? If so, then there is an error with your board's design/implementation.
Lastly, I am having a fairly good experience with my board. Perhaps my limited hearing (nothing over 2Khz) spares me from any deficiencies it might have. Hence, I needed my GF to help me "voice" the crossover between my sub channel (now modified to a 876hz linkwitz-riley LP filter) and my L-R channel drivers.
Since you seem very capable of modifying your board, I can determine values if you want to change the sub xover frequency (assuming its working correctly).
Last edited:
Those two capacitors are 1uF.
Just did a dirty messure and it´s to high for me.
The cross should be 80hz or even lower and I asume its first order cross.
Not so useful.
I can solder but not much more than that...🙂
Hi Matt, sorry for the radio silence.
I've not read all recent posts but thought I'd mention the following, in case this helps.
I always drive my amp (with input transformers) from a low impedance source. If you have a potentiometer (volume control) prior to the transformer primary, then the impedance of the source driving the transformer is probably high and varying with volume level. This will cause issues (noise, hum) with many input transformers. You can test this by cutting out the volume control and driving the input transformer directly from a low impedance source (a laptop headphone output would be fine for testing, with volume controlled in the digital domain).
Don't 'short' or leave open any section of the transformer primary. It sounds like you've connected the signal of your unbalanced source to the transformer primary + connection, ground to the CT and then played with having the primary - connection floating or tied to CT. Try connecting the source signal to the transformer primary + connection and source ground to the transformer primary - connection, so the whole primary coil is in series with the source.
You're correct that I've never connected the secondary CT to amp input ground, as I was concerned that this would provide a route to ground for the amp input bias voltage. I didn't have any problems with my amp board and transformer with this approach. Remember that the input signals will be connected to amp input ground anyway, as the amp inputs have an impedance to ground.
Good luck 🙂
Good point - to preserve the same Fc, the input caps (C1,2 in the schematic you posted) should be reduced by a factor of 10, to 100nF.
The ones I intended be swapped were the feedback resistors (R2,3) to preserve unity gain.
Which resistors on your schematic do the 33ks correspond to?
Your analysis seems to me to be correct. The idea behind my suggestion was that harshness comes about from noisy power supplies (in general) and the 4k7s put a load on the power supply of whatever is driving this amp. A heavy load induces harshness in the source component, via its power supplies.
More than happy to oblige 🙂
Those caps could be the decouplers for the opamp supplies, yes. In which case they could contribute to the harshness by coupling noise from the two power rails (+ and -) into your GND. Whether they do in fact do that depends on the layout of the GND connections.
You could experiment by removing both and putting in an extra wire from the centre point of the pair - in essence a dedicated GND. You do this by leaving the cap connections to opamp pins 4 (-ve supply) and 8 (+ve supply) in place but don't restore the pin on each cap to the GND connection. Instead connect the two cap pins that were formerly going to GND together, then from that junction wire a GND connection to a suitable spot. Hard to determine what the best GND point to attach to without a more comprehensive schematic - you can experiment for best sound.
Oh, sorry, I meant the rectangular polyester black caps, not the round cans. The 1uf cans are simply bias resistors between op-amp stages (to mitigate any DC offset). The two polyester caps are the caps that determine Fc of the LP filter.
BTW: the two 4k7 resistors that you replace in the photo, one is in the LP filter circuit, the other is a DC bias resistor for the Op-amps.
The graph shows the LP with an Fc currently between 200 and 300 hz. I presume this is post your resistor modifications.
Can you take a photo, much like above, but showing both op-amps and the all 4 resistors you replaced?
Last edited: