Aah, you didit! Very cool!
Me too wanted to go smallish, but didn’t dare as I’m totally not fearless.
Kind of late in the day and maybe not so important, but anyone building this amp for the first time, I feel it should be recommended that a transformer with 20V secondaries is probably optimum but only marginally - you know, if you're an OCD type go with 20V secondaries on your transformer.
In all reality there is probably very little difference, provided you have chosen the right shunt references etc.
Out of all the first watt amps we have to play with, I think this one could certainly be recommended with using a transformer with 20V secondaries, or at least the general recommendation should be that a transformer with secondary voltages between 18V and 20V should be used.
In all reality there is probably very little difference, provided you have chosen the right shunt references etc.
Out of all the first watt amps we have to play with, I think this one could certainly be recommended with using a transformer with 20V secondaries, or at least the general recommendation should be that a transformer with secondary voltages between 18V and 20V should be used.
Why?
I’m not disagreeing, but that’s an awful lot of words with no reason or information behind it, aside from soothing possible OCD...
I’m not disagreeing, but that’s an awful lot of words with no reason or information behind it, aside from soothing possible OCD...
Mostly because of the shunt reference issue.
If you went and inspected 50 F4 amps that were built 5 or more years ago, I guarantee 70% aren't even functioning as intended.
You would also be providing the mosfets some additional voltage to be operating away from the non-linear capacitance region of the mosfet.
The amp doesn't have negative feedback so all these small things are worthwhile.
If you went and inspected 50 F4 amps that were built 5 or more years ago, I guarantee 70% aren't even functioning as intended.
You would also be providing the mosfets some additional voltage to be operating away from the non-linear capacitance region of the mosfet.
The amp doesn't have negative feedback so all these small things are worthwhile.
Last edited:
I've got some of the recommended 431's on hand (NCP431ACLPRAG), just need to get in there and swap them out. I use 20V secondaries, but that was just to make up for some lost voltage to juma's cap multiplier. Rails are 22.7V on mine, I think. Likely leaving some performance on the table as it sits now...
he's Pico
unable to stop his own OCD at green leds, as is case with me
Now what are you up to?
FPGA biasing circuit?
Or something way more dumb but even more cunning?
Field programmable gate array chip. I've never done anything with them but they are kind of cool devices to mess around with.
naah, I mean L'Art pour L'Art approach is always interesting, but I have other fields of interest for same 🙂
431, when used proper one is perfectly doing it's job
though, as you already know, there are plenty of possibilities to replace it, be it complicated or not ..... LM336Z-5 (two of them in series, stable up down to minuscule uA) ...... if some more voltage is needed , adding in series something as you did with sissy F6 - I mean something with negative TempCo - sissy LED or even 1N4148
two pieces of 431 will give 10V biasing window for gate-to-gate ......... adding 1 or 2pcs of 1N4148 will give 0V5 per each, even at flimsy 100uA
however , choosing appropriate 431 is more than proper solution, needing 30min o rummaging through datasheets
in any case , I understand Papa's approach and situation - he had proper TL431 in drawer and he most probably even didn't think that there are inappropriate ones on the market
been there, done that, paid dearly
431, when used proper one is perfectly doing it's job
though, as you already know, there are plenty of possibilities to replace it, be it complicated or not ..... LM336Z-5 (two of them in series, stable up down to minuscule uA) ...... if some more voltage is needed , adding in series something as you did with sissy F6 - I mean something with negative TempCo - sissy LED or even 1N4148
two pieces of 431 will give 10V biasing window for gate-to-gate ......... adding 1 or 2pcs of 1N4148 will give 0V5 per each, even at flimsy 100uA
however , choosing appropriate 431 is more than proper solution, needing 30min o rummaging through datasheets
in any case , I understand Papa's approach and situation - he had proper TL431 in drawer and he most probably even didn't think that there are inappropriate ones on the market
been there, done that, paid dearly
Output Mosfet's order?
Hi friends
I've got 2 sets of DIYAudiostore's Output Mosfets.
They've got numbers on them which, according to the description, are not related to their matching/measurement…
So I conclude that their respective order and pairing don't matter? (like, put the lower numbers N-channel / P-channel into board 1, the higher numbers into board 2?
(It's even a little more complicated: I've got 2 sets (4 boards)...
thanks again!
david
Hi friends
I've got 2 sets of DIYAudiostore's Output Mosfets.
They've got numbers on them which, according to the description, are not related to their matching/measurement…
So I conclude that their respective order and pairing don't matter? (like, put the lower numbers N-channel / P-channel into board 1, the higher numbers into board 2?
(It's even a little more complicated: I've got 2 sets (4 boards)...
thanks again!
david
Are the numbers on the front (plastic) or the back (metal)?
If on the plastic, those are arbitrary. Don’t separate the trios from themselves.
If on the metal, that’s the Vgs, which is the actual measurement. Still, don’t separate the trios.
And to answer the next question, the Vgs of the N channel will almost never match the Vgs of the P channel, and they don’t need to for proper operation.
If on the plastic, those are arbitrary. Don’t separate the trios from themselves.
If on the metal, that’s the Vgs, which is the actual measurement. Still, don’t separate the trios.
And to answer the next question, the Vgs of the N channel will almost never match the Vgs of the P channel, and they don’t need to for proper operation.
FPGA = Field Programmable Gate Array
Sufficiently complex that one might expect the Mighty ZM to use it in a biasing circuit. Others might use it for speaker protection as well 😉
Sufficiently complex that one might expect the Mighty ZM to use it in a biasing circuit. Others might use it for speaker protection as well 😉
I've built a few tube kits and I just finished two Amp Camps and a B1 Korg. They work and sound great.
I think I want to build this amplifier, but I've never built anything from "scratch". I've read through some of this thread and the F4 thread, but haven't digested every post...yet.
I know what a schematic looks like and can identify most parts, but I am not an electrical engineer. I have a MM and a small scope, and I'd like to learn more, but to be honest, this will not be my life's work. I like good audio, I don't like spending gobs of money on it. I like projects, but I don't like projects that I can't finish.
I think the F4 is a good fit. I have a Grounded Grid preamp with the following specs:
Harmonic Distortion: Less than .02% at 2 volts RMS out.
Signal to Noise Ratio: Greater than 93 dB.
Bandwidth: 1 dB down from 5 Hz to 300 kHz.
Gain: 12 dB.
Max Signal Out: 20 volts RMS.
Output Impedance: 200 ohms.
Input Impedance: 50 k Ohms.
Inputs: 3, Outputs: 1, all single-ended, RCA
Controls: Input Selector, Volume, Power
Tube Compliment: 3-12AU7
Dimensions Overall: 15'W by 11'D by 4 1/4'H.
100-120V, 200-240V, 50-60 Hz, 25 watts
I'd likely be trying this F4 on a pair of Altec A5 VOTT that I built 8-9 years ago with the Hiraga crossovers, replacing a 3W/ch TS cathode follow tube amp. The A5s are about 98dB sensitive. I also have 2 pair of Altec 604Cs that need power too. Point is, I like lower power amps and efficient speakers.
So all that said, other than ordering the F4 and PSU boards, the Deluxe Aluminum 4U chassis, some Keratherm insulators and everything on the BoM, what do I need? I guess I'll need binding posts, IEC and RCA sockets, wire, etc. Am I biting off too much?
I think I want to build this amplifier, but I've never built anything from "scratch". I've read through some of this thread and the F4 thread, but haven't digested every post...yet.
I know what a schematic looks like and can identify most parts, but I am not an electrical engineer. I have a MM and a small scope, and I'd like to learn more, but to be honest, this will not be my life's work. I like good audio, I don't like spending gobs of money on it. I like projects, but I don't like projects that I can't finish.
I think the F4 is a good fit. I have a Grounded Grid preamp with the following specs:
Harmonic Distortion: Less than .02% at 2 volts RMS out.
Signal to Noise Ratio: Greater than 93 dB.
Bandwidth: 1 dB down from 5 Hz to 300 kHz.
Gain: 12 dB.
Max Signal Out: 20 volts RMS.
Output Impedance: 200 ohms.
Input Impedance: 50 k Ohms.
Inputs: 3, Outputs: 1, all single-ended, RCA
Controls: Input Selector, Volume, Power
Tube Compliment: 3-12AU7
Dimensions Overall: 15'W by 11'D by 4 1/4'H.
100-120V, 200-240V, 50-60 Hz, 25 watts
I'd likely be trying this F4 on a pair of Altec A5 VOTT that I built 8-9 years ago with the Hiraga crossovers, replacing a 3W/ch TS cathode follow tube amp. The A5s are about 98dB sensitive. I also have 2 pair of Altec 604Cs that need power too. Point is, I like lower power amps and efficient speakers.
So all that said, other than ordering the F4 and PSU boards, the Deluxe Aluminum 4U chassis, some Keratherm insulators and everything on the BoM, what do I need? I guess I'll need binding posts, IEC and RCA sockets, wire, etc. Am I biting off too much?
Last edited:
The F4 is my favorite Firstwatt amp, and well worth the effort. Just a couple notes about parts... The transistor kit from the Diyaudio store gives you the 431's, but I would purchase the part # I listed in post #2267 to make sure things work properly. I prefer the Harris P-channels but I don't know if there's anyone offering matched triplets right now, you would have to do some detective work and ask around. It's not a make or break deal though by any means. I needed to change R9 to 5k to get the bias current where I wanted it. The Linear Systems JFET's from the store should work great. I used some matched BL grade Toshiba's from my stash. Very likely to get fakes trying to buy those guys these days though.