Speaking of bias...
I have ran mine as high as 1.25V bias (with both 25V and 30V rails). The heatsinks are still fairly cool (I can hold into them indefinitely, maybe 50 degrees C)...What bias are you guys running?
I have ran mine as high as 1.25V bias (with both 25V and 30V rails). The heatsinks are still fairly cool (I can hold into them indefinitely, maybe 50 degrees C)...What bias are you guys running?
Pass DIY Addict
Joined 2000
Paid Member
Andrew: since the LDRs will feed directly into the BA-3 and the input impedance for the BA-3 is high enough at 47k, i didn't think I'd need the buffer/follower. It seems that this is functionally equivalent to putting a pot in front of the BA-3 for volume control. The relays from the dCB1 will serve for input selection.
Hikari: 21v is what I'll get from my 20v transformer, once rectified and then the 7v loss that Salas cites for the regulator. I think I'll switch transformers with the 28v one that I have for my Pearl project. This will give me more room to push rail voltage up with the DCB1 as a regulator, and then I won't need the extra power resistors in my Pearl PSU. I should be able to hit 30v with ease by adjusting the resistor in series with the voltage LEDs.
Adjusting the current with the DCB1 regulator is no trouble, I bolted the mosfets to the bottom of the chassis so they should handle the heat well. I'll go back to the BA-3 article to find the bias levels. With my prev configuration for the DCB1, it was providing about 350mA. I'll have to recheck with new voltage levels.
Hikari: 21v is what I'll get from my 20v transformer, once rectified and then the 7v loss that Salas cites for the regulator. I think I'll switch transformers with the 28v one that I have for my Pearl project. This will give me more room to push rail voltage up with the DCB1 as a regulator, and then I won't need the extra power resistors in my Pearl PSU. I should be able to hit 30v with ease by adjusting the resistor in series with the voltage LEDs.
Adjusting the current with the DCB1 regulator is no trouble, I bolted the mosfets to the bottom of the chassis so they should handle the heat well. I'll go back to the BA-3 article to find the bias levels. With my prev configuration for the DCB1, it was providing about 350mA. I'll have to recheck with new voltage levels.
Last edited:
Andrew: since the LDRs will feed directly into the BA-3 and the input impedance for the BA-3 is high enough at 47k, i didn't think I'd need the buffer/follower. It seems that this is functionally equivalent to putting a pot in front of the BA-3 for volume control. The relays from the dCB1 will serve for input selection.
Hikari: 21v is what I'll get from my 20v transformer, once rectified and then the 7v loss that Salas cites for the regulator. I think I'll switch transformers with the 28v one that I have for my Pearl project. This will give me more room to push rail voltage up with the DCB1 as a regulator, and then I won't need the extra power resistors in my Pearl PSU. I should be able to hit 30v with ease by adjusting the resistor in series with the voltage LEDs.
Adjusting the current with the DCB1 regulator is no trouble, I bolted the mosfets to the bottom of the chassis so they should handle the heat well. I'll go back to the BA-3 article to find the bias levels. With my prev configuration for the DCB1, it was providing about 350mA. I'll have to recheck with new voltage levels.
Yeah, you could do that. The shunt might get too hot, depending on your chassis. The Salas shunt in my phono stage gets plenty hot...
An alternative is you could build the DCB1 with all the stuff you have and do a Ba-3 your 20v transformer and a LM317/337 which only needs to drop 3 volts or so. That would get you 24-ish volt rails. I haven't found the BA-3 to be too influenced by regulators....
The Mesmerize uses either 5V or 12V relays fed from the 12V 3pin chip regulator.
The LDRs could be fed from a 5V 3pin chip regulator.
Using a DCB1 as a regulator for these two duties is a complete waste.
The LDRs could be fed from a 5V 3pin chip regulator.
Using a DCB1 as a regulator for these two duties is a complete waste.
Good news! GlassWare has a new suitable PSU for this project - the PS-19. This seems to be the replacement for the PS-12 I used in my build.
GlassWare PS-19 Bipolar Power Supply Kit
GlassWare PS-19 Bipolar Power Supply Kit
Those Glassware boards are all ridiculously nice. Very well thought out, and extremely well made.
Received my BA-3 pcb and I'm about to begin stuffing the board. In the meantime I tried to find proper transformer VA rating for BA-3 gain stage,couldn't find any info about it.
Any suggestion?
Any suggestion?
It's all in the article: frontend around 8mA, MOSFETs biased at 45 mA.
So 0.053 x 24 V = 1.3 Watt. x2 for positive and negative half. So 3 Watt is pulled from the circuit. So a minimum 6 VA transformer will do.
Though I would use a bit bigger trafo of 10 or 15 VA
So 0.053 x 24 V = 1.3 Watt. x2 for positive and negative half. So 3 Watt is pulled from the circuit. So a minimum 6 VA transformer will do.
Though I would use a bit bigger trafo of 10 or 15 VA
Hi all
I am in multiway, going out a digital XO - DAC - Preamp with volume control (Najda board on DIYA) and feeding F4s. The Najda output can be selected from 2v to 6v, I think it is better to stay at 2v which is the standard level.
Thus I will stuff one BA3 in each F4 case with the same PSU, as per 6L6 advice.
I have two questions :
1- how to adapt levels to use all F4 swing without saturation and distorsion and without risk for compressions drivers. Should I add a scale of resistors before or just change input resistors R1 and R2 (probably increasing R1). I guess I need to go from 2v to 2x20=40, so a x20 factor instead of the x30 stock gain.
2- I use FQP3N30 and FQP3P20 Fairchild devices with 3 to 5v Vgs so I understand I have to increase P1 and P2 to 1k, or I may try with the 500 first ?
Thanks a lot
Jean-Louis
I am in multiway, going out a digital XO - DAC - Preamp with volume control (Najda board on DIYA) and feeding F4s. The Najda output can be selected from 2v to 6v, I think it is better to stay at 2v which is the standard level.
Thus I will stuff one BA3 in each F4 case with the same PSU, as per 6L6 advice.
I have two questions :
1- how to adapt levels to use all F4 swing without saturation and distorsion and without risk for compressions drivers. Should I add a scale of resistors before or just change input resistors R1 and R2 (probably increasing R1). I guess I need to go from 2v to 2x20=40, so a x20 factor instead of the x30 stock gain.
2- I use FQP3N30 and FQP3P20 Fairchild devices with 3 to 5v Vgs so I understand I have to increase P1 and P2 to 1k, or I may try with the 500 first ?
Thanks a lot
Jean-Louis
You can try the 500k...you may not be able to bias it up without maxing out the pot. Probably safer to start with 1k so you dont have to desolder the 500k pot if it doesnt work. You could also add a resistor to the 500k pot but that might require cutting a trace on the pcb.
Can I use 220 ohms, 3W instead of 330 ohms at output
i do not have 330 ohms, 3W resistor that is in the circuit at the output before the output capacitor. Can I use a 220 ohms, 3W instead. I do have a good quality Panasonic 220 ohms in stock and cannot get decent ones in the local market here. BTW what is the purpose of that resistor? Is it to clamp the output to some resistive value?
Regards
i do not have 330 ohms, 3W resistor that is in the circuit at the output before the output capacitor. Can I use a 220 ohms, 3W instead. I do have a good quality Panasonic 220 ohms in stock and cannot get decent ones in the local market here. BTW what is the purpose of that resistor? Is it to clamp the output to some resistive value?
Regards
You will have slightly lower gain than with the stock value of 330R.
Thanks for the advice.
It doesn't have to be that way. Gain in this circuit is a function of MOSFET's transconductance which raises with current. So raise the current through MOSFETS to 100mA and you'll have plenty of gain (if heatsinks allow that).You will have slightly lower gain than with the stock value of 330R.
- Home
- Amplifiers
- Pass Labs
- The BA-3 as preamp build guide