Im in the process of setting up my amp. It has 2 amps.
One based on the Rod Elliot P3A and the other is based on the TDA7294.
The speakers are
1. A pair of Eminence Gama 15 Inch woofers. (In a massive horn cabinet).
2. One pair of 8 inch full range speakers.
3. One pair silk dome tweeters peerless.
4. One pair Eminence CD 2002.
The idea is to split them between the the two amps.
On testing I found that the TDA7294 gain is crazy (Max value as per data sheet). My bluetooth volume goes from 1 to 30.
I can not play even 2 on the TDA without it sounding too loud.
On the P3A I can take the volume upto max in short bursts.
This forces me to make a call.
Do I reduce the gain on the TDA or increase the Gain on the p3a. (Finally there will be a balance control anyway). Reducing the gain on the TDA seems easy and less complicated to me.
My question is will reducing the gain improve sound quality i.e. lower distortion. My exp with old amps is the slower, lower HFE transistors with low gain seem to work better i.e. less distortion and very clean sound.
Sorry for the Noob Question.
Both boards are being powered from the same transformer. Separate PSU boards. 35-0-35 DC @ 15 amps.
Any tips on how to increase the gain on the P3A. i.e. which stage ?. Or all stages.
One based on the Rod Elliot P3A and the other is based on the TDA7294.
The speakers are
1. A pair of Eminence Gama 15 Inch woofers. (In a massive horn cabinet).
2. One pair of 8 inch full range speakers.
3. One pair silk dome tweeters peerless.
4. One pair Eminence CD 2002.
The idea is to split them between the the two amps.
On testing I found that the TDA7294 gain is crazy (Max value as per data sheet). My bluetooth volume goes from 1 to 30.
I can not play even 2 on the TDA without it sounding too loud.
On the P3A I can take the volume upto max in short bursts.
This forces me to make a call.
Do I reduce the gain on the TDA or increase the Gain on the p3a. (Finally there will be a balance control anyway). Reducing the gain on the TDA seems easy and less complicated to me.
My question is will reducing the gain improve sound quality i.e. lower distortion. My exp with old amps is the slower, lower HFE transistors with low gain seem to work better i.e. less distortion and very clean sound.
Sorry for the Noob Question.
Both boards are being powered from the same transformer. Separate PSU boards. 35-0-35 DC @ 15 amps.
Any tips on how to increase the gain on the P3A. i.e. which stage ?. Or all stages.
to increase the gain on the P3A, increase R5
beware that changes to gain will/may affect dc offset, stability, thd etc
beware that changes to gain will/may affect dc offset, stability, thd etc
"Do I reduce the gain on the TDA or increase the Gain on the p3a. (Finally there will be a balance control anyway). Reducing the gain on the TDA seems easy and less complicated to me."
I would do neither and fit a preset volume control to the input of the TDA.
That will save screwing up the characteristics of the TDA.
I would do neither and fit a preset volume control to the input of the TDA.
That will save screwing up the characteristics of the TDA.
John I think changing the gain on the TDA will not screw up anything if anything reducing it may improve things. Is my thinking and what initall testing is showing. Also as the TDA seems to be the more powerfull of the two Im thinking it should be assigned to driving the large woofers. I will try your idea. Also.
ctrlx thanks will try that. What I don't understand is why even after setting the gain to the same values say P3a circuit says 27 db and i match the TDA to 27 one still sounds louder than the other.
This is what chatgpt advises.
=============
In the P3A amplifier design, the gain can be adjusted by changing the values of the input resistors connected to the non-inverting and inverting inputs of the amplifier. The gain formula for this type of amplifier is:
Gain=Rf/Rin
Where:
Commonly used values for the P3A amplifier might include feedback resistors in the range of 22kΩ to 47kΩ and input resistors in the range of 1kΩ to 10kΩ. The gain can be adjusted according to your requirements by selecting appropriate resistor values.
To determine the gain of your P3A amplifier, you should refer to the documentation or schematic provided by Rod Elliott or the specific design resources you are using for your build. Additionally, you can use the gain formula mentioned above to calculate the gain based on the resistor values you choose for your specific configuration.
(Yes I know some stuff did not copy paste over right from chatgtp have tried to fix it).
ctrlx thanks will try that. What I don't understand is why even after setting the gain to the same values say P3a circuit says 27 db and i match the TDA to 27 one still sounds louder than the other.
This is what chatgpt advises.
=============
In the P3A amplifier design, the gain can be adjusted by changing the values of the input resistors connected to the non-inverting and inverting inputs of the amplifier. The gain formula for this type of amplifier is:
Gain=Rf/Rin
Where:
- Gain is the voltage gain of the amplifier.
- Rf is the resistance of the feedback resistor connected between the amplifier's output and the inverting input.
- Rin is the resistance of the input resistor connected to the non-inverting input.
Commonly used values for the P3A amplifier might include feedback resistors in the range of 22kΩ to 47kΩ and input resistors in the range of 1kΩ to 10kΩ. The gain can be adjusted according to your requirements by selecting appropriate resistor values.
To determine the gain of your P3A amplifier, you should refer to the documentation or schematic provided by Rod Elliott or the specific design resources you are using for your build. Additionally, you can use the gain formula mentioned above to calculate the gain based on the resistor values you choose for your specific configuration.
(Yes I know some stuff did not copy paste over right from chatgtp have tried to fix it).
Last edited:
Tda datasheet defines a min gain of 24 dB (less could be possible, but stability is not guaranteed). Max is 40 dB. I would not touch p3a gain unless officially described in rod elliotts project.
you will need a level potentiometers anyway to adjust for speaker sensitivities.
You may not have calculated gain correctly or your speakrs differ in sensitivity.
Also have a look at this thread regarding gain structure:
https://www.diyaudio.com/community/threads/what-is-gain-structure.186018/
you will need a level potentiometers anyway to adjust for speaker sensitivities.
That is really strange.
You may not have calculated gain correctly or your speakrs differ in sensitivity.
Also have a look at this thread regarding gain structure:
https://www.diyaudio.com/community/threads/what-is-gain-structure.186018/
I am testing both amps using the same speakers.
Ok some more TDA7294 related questions.
I am hearing a power on pop and a power off pop.
My present values are 10k between mute and +
And 22k with 10uf for stand by. My plan is to increase the 10uf to 47uf will it help introduce a delay any other ideas.
They say the power off pop is not dangerous but Id rather not hear it. The board has 4 x 4700 uf caps.
My P3A board has more so when I power off the music plays for a few seconds and fades off which is fine. There is a power on pop though. But only sometimes.
For reference here is a schematic.
I will go back to measuring i.e. send in 1 volt peak to peak and measure the output from each amp so I have exact voltage numbers. I will match those instead of trying to play with the gain.
Problem is I have moved everything off my workbench onto one of the large speakers will need to move it all back to my workbench which is already full up with other work.
ps: The input impedance on both amps is different maybe this is causing the issue. Best bet is to just match voltage in vs voltage out. For both. Still have one STK 100 watt board to test. Then out of the 3 will pick 2. Worst case pick the best and just build one more of that.
Ok some more TDA7294 related questions.
I am hearing a power on pop and a power off pop.
My present values are 10k between mute and +
And 22k with 10uf for stand by. My plan is to increase the 10uf to 47uf will it help introduce a delay any other ideas.
They say the power off pop is not dangerous but Id rather not hear it. The board has 4 x 4700 uf caps.
My P3A board has more so when I power off the music plays for a few seconds and fades off which is fine. There is a power on pop though. But only sometimes.
For reference here is a schematic.
I will go back to measuring i.e. send in 1 volt peak to peak and measure the output from each amp so I have exact voltage numbers. I will match those instead of trying to play with the gain.
Problem is I have moved everything off my workbench onto one of the large speakers will need to move it all back to my workbench which is already full up with other work.
ps: The input impedance on both amps is different maybe this is causing the issue. Best bet is to just match voltage in vs voltage out. For both. Still have one STK 100 watt board to test. Then out of the 3 will pick 2. Worst case pick the best and just build one more of that.
Attachments
Last edited:
just a remark: if the GIF shows the values of your circuit the amp has around 32 dB gain, not 27 dB.
you would need to reduce the feedback resistor to around 15 kOhm (slightly less or you dial in a higher value to ground with your 100 ohm trimmer) for 27 dB.
you would need to reduce the feedback resistor to around 15 kOhm (slightly less or you dial in a higher value to ground with your 100 ohm trimmer) for 27 dB.
I have no idea why the perceived loudness differs between these amps when set to the same gain, unless one of them has a fault or is oscillating. However a simple way to adjust the sensitivity is just to add a simple two resistor attenuator to the input of the louder one.
You could start with a small 10k trimmer and turn it until the levels are equal and then substitute fixed resistors.
If you can supply a known steady signal from a tone generator and measure input and output levels you would then be able to get to the right values by simple calculation.
If you wish to experiment you could try say a 6.8K in series with 3.3K and see if you need more or less attenuation. Try 4.7K and 4.7K if that drops the signal too much. I’d try to keep the series total value to around 10K as that shouldn’t be too low if you are driving it from opamps and this means the 22K input impedance won’t put your calculation out too much.
At some point you’ll need some way of measuring all this anyway.
Best of luck with this.
You could start with a small 10k trimmer and turn it until the levels are equal and then substitute fixed resistors.
If you can supply a known steady signal from a tone generator and measure input and output levels you would then be able to get to the right values by simple calculation.
If you wish to experiment you could try say a 6.8K in series with 3.3K and see if you need more or less attenuation. Try 4.7K and 4.7K if that drops the signal too much. I’d try to keep the series total value to around 10K as that shouldn’t be too low if you are driving it from opamps and this means the 22K input impedance won’t put your calculation out too much.
At some point you’ll need some way of measuring all this anyway.
Best of luck with this.
No the gif was putup just to show my mute and standby situation / values.
I have read the data sheets and have been playing with these values but it doesn't seem to change much.
Technically this setup should not have a turn on thump or a turn off pop. But it does.
I have a function generator and a scope so will use that to balance out the gain. But question remains is there any correlation between gain and distortion. Guess Ill just measure it.
I did thy changing the cap between pin 6 and pin 14 from 22uf to 47uf was a bit disappointed with the drop in clarity. And no improvement to bass. Im trying all the mods. One by one.
Thank you for your inputs and ideas.
Ps: I over came the issue of my body / presence impacting the distortion figures by shielding everything using copper tape. And the highest quality shielded wires you get here.
I do have access to cat 5 and cat 7 network cables. Im wondering if I should start using that.
Crimping the tiny wires is proving quite hard. The cat cables have thicker wires. And are easier to crimp even to the smallest of connectors.
I have read the data sheets and have been playing with these values but it doesn't seem to change much.
Technically this setup should not have a turn on thump or a turn off pop. But it does.
I have a function generator and a scope so will use that to balance out the gain. But question remains is there any correlation between gain and distortion. Guess Ill just measure it.
I did thy changing the cap between pin 6 and pin 14 from 22uf to 47uf was a bit disappointed with the drop in clarity. And no improvement to bass. Im trying all the mods. One by one.
Thank you for your inputs and ideas.
Ps: I over came the issue of my body / presence impacting the distortion figures by shielding everything using copper tape. And the highest quality shielded wires you get here.
I do have access to cat 5 and cat 7 network cables. Im wondering if I should start using that.
Crimping the tiny wires is proving quite hard. The cat cables have thicker wires. And are easier to crimp even to the smallest of connectors.
The presence of distortion that varies as you move your hand near the circuit is an indication of oscillation. My first TDA7293 did that when I had wires everywhere on my bench. I don’t know what frequency it was as I didn’t have a scope at the time.
I was checking dc offset and my digital meter read 0V, but my analogue meter read various values according to where I put my hand! Tidying up the wiring with screened or twisted wires helped as you mentioned.
By the way the mute circuit on the data sheet should work ok to prevent power on pops, but unless you have a control signal that changes level when the power drops, it won’t prevent the power off pop or squeal.
The gif above has nothing connected to mute/standby pins.
A mute/dc protect relay is a better option although more complex and expensive.
I don’t think changing the gain on the tda7294 will audibly change the distortion, although you may be able to measure the change.
I hear you about the crimp connectors! It took me a long while to get used to them and even now I don’t get it right every time.
I was checking dc offset and my digital meter read 0V, but my analogue meter read various values according to where I put my hand! Tidying up the wiring with screened or twisted wires helped as you mentioned.
By the way the mute circuit on the data sheet should work ok to prevent power on pops, but unless you have a control signal that changes level when the power drops, it won’t prevent the power off pop or squeal.
The gif above has nothing connected to mute/standby pins.
A mute/dc protect relay is a better option although more complex and expensive.
I don’t think changing the gain on the tda7294 will audibly change the distortion, although you may be able to measure the change.
I hear you about the crimp connectors! It took me a long while to get used to them and even now I don’t get it right every time.
If your circuit looks like this, replace R3 with a 10 kΩ resistor for 23.9 dB gain. You can also put another 22 kΩ resistor in parallel with R3. That'll give you a total of 11 kΩ, which results in 24.7 dB gain.
Changing the gain will allow you to go up to maybe 3-4 on your 0-30 scale. The best solution is to add a volume control.
Tom
Changing the gain will allow you to go up to maybe 3-4 on your 0-30 scale. The best solution is to add a volume control.
Tom