The search term you need is "Gain clone".
Here is one.
Tube Gainclone for DIY Buffs
Several other search results will direct you right back to DIYAudio threads.
Ian
Here is one.
Tube Gainclone for DIY Buffs
Several other search results will direct you right back to DIYAudio threads.
Ian
Ian, very interesting schematic (project)!
What changes to the schematic should I do instead to replace LM3875 with LM1875 or LM3886.
What tube/valve participate in this project.
Thanks for your help
Last edited:
Typically a ECC88 / 6DJ8 /6922 tube is used.
Supply voltages for the LM1875 are lower. The gain will have to be altered too I think.
Why do you want to add the tube ? The LM1875 sounds very good by itself. The tube will degrade signal to noise ratio and add more harmonics to the sound most likely making it sound 'warmer'. But some prefer that route ! Everyone gets what they want !
Supply voltages for the LM1875 are lower. The gain will have to be altered too I think.
Why do you want to add the tube ? The LM1875 sounds very good by itself. The tube will degrade signal to noise ratio and add more harmonics to the sound most likely making it sound 'warmer'. But some prefer that route ! Everyone gets what they want !
My simple answer would be , maybe you should use the LM1875 as it is shown in the application note for the chip. It works very well and has good sound.
Simple uncomplicated and good sound !
http://www.ti.com/lit/ds/symlink/lm1875.pdf
Simple uncomplicated and good sound !
http://www.ti.com/lit/ds/symlink/lm1875.pdf
Ian, Is can be tube/valve voltage increased to + /-45V (+ /-90V) and ate without any changes needed in the schematic (Tube Gainclone) hybrid amplifier with LM3875?
thank you
Had to "duck off" and have a look at the LM3875 datasheet and the circuit on the The jLTi site.
The power amp chip data says voltage rails between +/- 20V and +/- 84 volts.
I would aim for say +/- 60V that won't bother the ECC88 at all.
That 10K cathode resistor means that the ECC88 current with +/-35V is going to be just a little less than 3.5mA.
Going up to +/-60V rails means that ECC88 current goes up to just less than 6mA. ECC88 are often run at that current and even higher. It will probably make it sound a little less "tubey" as the ECC88 is then on a more linear part of the curve.
You could increase that 10K to get back to the 3.5 mA'ish operating point.
15K would suit about +/-50V rails
18K would suit about +/-65V rails
22K would suit about +/- 80 V rails
Personally I would not go above around +/- 65 to 70 Volts. Thats just me being conservative, why? because of lots of bad experiences when I was young and stupid trying to squeeze the last watt out of circuits and usually squeezing out charcoal instead.
Cheers,
Ian
The power amp chip data says voltage rails between +/- 20V and +/- 84 volts.
I would aim for say +/- 60V that won't bother the ECC88 at all.
That 10K cathode resistor means that the ECC88 current with +/-35V is going to be just a little less than 3.5mA.
Going up to +/-60V rails means that ECC88 current goes up to just less than 6mA. ECC88 are often run at that current and even higher. It will probably make it sound a little less "tubey" as the ECC88 is then on a more linear part of the curve.
You could increase that 10K to get back to the 3.5 mA'ish operating point.
15K would suit about +/-50V rails
18K would suit about +/-65V rails
22K would suit about +/- 80 V rails
Personally I would not go above around +/- 65 to 70 Volts. Thats just me being conservative, why? because of lots of bad experiences when I was young and stupid trying to squeeze the last watt out of circuits and usually squeezing out charcoal instead.
Cheers,
Ian
With higher voltages on the tube the only component that might need any change is the 10K/1W resistor on the cathode of the tube. +/- 45 should work fine. You might just need to check if the operating current produces the best sound for the supply voltages you will be using.
The original circuit used the 6922 as it can work at lower voltages and use the same supply as the chip.
The 3886 has additional terminals to be connected as compared to the 3875. The 3875 reputedly sounds better than the 3886. You'll find references to that on this forum.
The original circuit used the 6922 as it can work at lower voltages and use the same supply as the chip.
The 3886 has additional terminals to be connected as compared to the 3875. The 3875 reputedly sounds better than the 3886. You'll find references to that on this forum.
With the same power supply Ucc=+ / - 35V, LM3875 provides 56W/4ohm and LM3886 provides 68W / 4 ohm, or this is a mistake?
No mistake, that must be correct.
But it's just 0.8 dB less than the 3886. You will not be able to hear the difference ! For better sound it is certainly not worth worrying about .
Note that power ratings are given at a particular distortion rating. And some circuits could clip earlier than others with the same supply voltage. So having some difference in wattage rating isn't surprising.
But it's just 0.8 dB less than the 3886. You will not be able to hear the difference ! For better sound it is certainly not worth worrying about .
Note that power ratings are given at a particular distortion rating. And some circuits could clip earlier than others with the same supply voltage. So having some difference in wattage rating isn't surprising.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.