Hello again everybody. It's time for me to build another amp. The The Valve Wizard valvewizard pcb looks almost perfect for my needs. Almost except for the diodes. I just really don't want them in there. I'd like to eliminate them if possible. But first I need to understand them.
D1: I assume I can just replace this with a resistor of the appropriate value.
D2: I suspect that D2 is there to protect V3 during amp turn on and warm up. Is that correct? If so, Is there a way to get the protection without diodes? What happens if it's simply eliminated?
D3 and D4: I don't know what these do. They are labeled for output protection, but I don't know how.
I appreciate the help.
Also, does anyone recommend using different tubes for V1 and V2? for example, using a 12AX7 for V1 for the mu and then 6DJ8 for V2 for its special qualities. This may necessitate a change in the RIAA circuit yes?
D1: I assume I can just replace this with a resistor of the appropriate value.
D2: I suspect that D2 is there to protect V3 during amp turn on and warm up. Is that correct? If so, Is there a way to get the protection without diodes? What happens if it's simply eliminated?
D3 and D4: I don't know what these do. They are labeled for output protection, but I don't know how.
I appreciate the help.
Also, does anyone recommend using different tubes for V1 and V2? for example, using a 12AX7 for V1 for the mu and then 6DJ8 for V2 for its special qualities. This may necessitate a change in the RIAA circuit yes?
D1: Correct, but the LED will give lower distortion.
D2: Correct, and when V3 is hot and conducting, D2 falls out of the circuit as it is reverse biased. Without it, you risk destroying V3.
D3 and D4 clamp the output to prevent an initial 150V or so surge hitting the output. This isn't a big deal if you're using a tube stage after this, but if you use a transistor stage, you might very well blow it. I've blown an opamp with a similar spike. You can replace them with a switch that shorts the output to ground until the circuit is hot and running, or a time-delay relay. See
D2: Correct, and when V3 is hot and conducting, D2 falls out of the circuit as it is reverse biased. Without it, you risk destroying V3.
D3 and D4 clamp the output to prevent an initial 150V or so surge hitting the output. This isn't a big deal if you're using a tube stage after this, but if you use a transistor stage, you might very well blow it. I've blown an opamp with a similar spike. You can replace them with a switch that shorts the output to ground until the circuit is hot and running, or a time-delay relay. See
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
This is really where merlin should answer. I'll try with my view ; D1 is needed ( and a good method) to bet bias for a tube. Why choose a less good method ? D2 is needed. Read the description, then you will understand why! D3 & D4 is there to protect the following amp from switch transient, they
will have no affect on the sound. Remove them and enjoy your speakers dance !
Tubes ? Read the http://www.valvewizard.co.uk/Phono_PCB_BOM.pdf for the available choices. And to have luck with the build, buy NEW tubes from a vendor that offers warranty.
will have no affect on the sound. Remove them and enjoy your speakers dance !
Tubes ? Read the http://www.valvewizard.co.uk/Phono_PCB_BOM.pdf for the available choices. And to have luck with the build, buy NEW tubes from a vendor that offers warranty.
Technically … the back-to-back pair end up clamping at |VZ + VF| or 0.7 volts above VZ.
Just nit picking…
GoatGuy
I admit my requirement for no diodes or transistors is fairly unreasonable. But that's where I am.
I take it you are avoiding the use of a STAND-BY switch ?? And using SS diodes..???
Even if you solve the Cathode to Heater issue.... Are all the other components able to withstand full B+ before the tubes are fully conducting???
For example: Some amps make over 500V B+ with no load...Then the 400V coupling caps and some 450V electrolytics will get stressed for about 10 seconds or so before voltages settle down under load... Just saying ...
Even if you solve the Cathode to Heater issue.... Are all the other components able to withstand full B+ before the tubes are fully conducting???
For example: Some amps make over 500V B+ with no load...Then the 400V coupling caps and some 450V electrolytics will get stressed for about 10 seconds or so before voltages settle down under load... Just saying ...
The output zeners are a good idea.
As was previously stated, it protects the following amp during turn on transient.
But here is something else that the zeners protect: I have not looked at the schematic, but I can guess . . it has an output coupling cap, right?
A coupling cap has voltage charge across it. When the amp is turned off, and the power supply voltage collapses, the charge across that coupling cap has to go somewhere, right? How about putting that into the other zener, not the following amp.
Most zener diodes do start to conduct at lower voltages than their rated zener voltage (early conduction effect). That is why I prefer to use a signal diode in series with each zener. Set 1, zener anode connected to signal diode anode. Set 2, zener cathode connected to signal diode cathode. Now you have a very high impedance for positive and negative music signals; there is no early conduction effect.
As was previously stated, it protects the following amp during turn on transient.
But here is something else that the zeners protect: I have not looked at the schematic, but I can guess . . it has an output coupling cap, right?
A coupling cap has voltage charge across it. When the amp is turned off, and the power supply voltage collapses, the charge across that coupling cap has to go somewhere, right? How about putting that into the other zener, not the following amp.
Most zener diodes do start to conduct at lower voltages than their rated zener voltage (early conduction effect). That is why I prefer to use a signal diode in series with each zener. Set 1, zener anode connected to signal diode anode. Set 2, zener cathode connected to signal diode cathode. Now you have a very high impedance for positive and negative music signals; there is no early conduction effect.
Thanks for all the info guys. I just ordered 2 pcbs. I can't decide on which tubes to try, so I'll try it 2 different ways. I'll replace the first diode with an appropriate resistor, the second diode with a neon lamp, and I'll just eliminate the last two since I'm going into tubes. I appreciate all the advice.
neon lamp is 50 - 60v before conduction, I would use 1N4148 diodes instead of the 1n4007. They are very fast and the 100v reverse voltage will act similar to neon lamp. The Split RIAA setup will reduce preamp hiss. btw, tubes and SS can be much better then tubes alone. I have a tidy 135v voltage tripler regulator for dual 6922 tubes that passes hum at levels lower then -100 db. (below the tube johnson noise)
Last edited:
heater supply
Hello,
I've build this amp as my first diy project. It's the only non-integrated phono I've heard and I'm not an expert, but in my point of view (compared to my R2R DAC for example) I'm very satisfied with the sound.
However when Merlin recommends using regulated heater supply, he really means it (at least with my mains and my tubes). I was fighting the power supply design and hum a lot until I used regulated supply.
So if you want to avoid silicon be prepared to implement good heater power supply.
Good luck with the project.
Hello,
I've build this amp as my first diy project. It's the only non-integrated phono I've heard and I'm not an expert, but in my point of view (compared to my R2R DAC for example) I'm very satisfied with the sound.
However when Merlin recommends using regulated heater supply, he really means it (at least with my mains and my tubes). I was fighting the power supply design and hum a lot until I used regulated supply.
So if you want to avoid silicon be prepared to implement good heater power supply.
Good luck with the project.
This is said quite often but it is simply not true. Datasheet for a yellow LED from Kingbright linked below. It starts conducting from ~1.65V and it is within specifications up until it has 20mA going through it which should be around 2.1V. A 12AX7 is going to have tons of distortion due to its position within the conduction curve of that LED. A 6922 will get you into the linear portion of the curve but it's most definitely not flat and will affect output.
https://www.kingbrightusa.com/images/catalog/SPEC/WP1513YT.pdf
from 1.8 to 2.1 v (0.3v) the current changes from 2 to 20ma = 17 ohms. down from 330 for 6 - 7 ma bias.
if signal levels are low the linearity is better. Signal at the grid appears at the cathode as a bias +- signal voltage, any variance in cathode load is then multiplied by the circuit gain.
positive half is muted, negative half is exaggerated. The sound produced was surprisingly good but somewhat muddy.
if signal levels are low the linearity is better. Signal at the grid appears at the cathode as a bias +- signal voltage, any variance in cathode load is then multiplied by the circuit gain.
positive half is muted, negative half is exaggerated. The sound produced was surprisingly good but somewhat muddy.
Feel free to do whatever you like. No need to listen to people who know what
they are doing. One learns by mistakes!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.