Baked, biased, and currently pumping some Nightwish! The BA-3/F-4 sounds awesome and, despite rather efficient speakers (Martin Logan 35XT; 92dB/4R), I'm not used to having to crank the volume to 3-o'clock to get rocking volume levels! Thanks all!
(I also was not expecting to have something of a space heater in my office! I know, I know, Class A lol)
(I also was not expecting to have something of a space heater in my office! I know, I know, Class A lol)
having it to 3-o'clock is good
read: https://www.diyaudio.com/community/threads/what-is-gain-structure.186018/
read: https://www.diyaudio.com/community/threads/what-is-gain-structure.186018/
I guys,
A few years ago I ordered the parts to build an F-4.
At that time the store shipped the International Rectifier (IR) version of the IRFP9140.
I’m currently looking to order the parts to build a second F-4 to run them in a Monoblock configuration.
However at the moment the store is shipping the Harris version of the IRFP9140.
If one Monoblock is based on Harris mosfets and the other on IR’s do you expect any sonic differences between the two?
Thanks in advance.
A few years ago I ordered the parts to build an F-4.
At that time the store shipped the International Rectifier (IR) version of the IRFP9140.
I’m currently looking to order the parts to build a second F-4 to run them in a Monoblock configuration.
However at the moment the store is shipping the Harris version of the IRFP9140.
If one Monoblock is based on Harris mosfets and the other on IR’s do you expect any sonic differences between the two?
Thanks in advance.
Hello there!
It took me a few months to report back
I got busy getting a DAC and building a streamer, just tired of records being so expensive, plus lack of space plus wanting a good-sounding option for exploring new music...
The F4 works and it's great! I tested it with my 94db speakers and my ultra-low power SET and realized that the SET simply doesn't have it. enough voltage gain. It works better with the 2a3 triode, but I never liked the sound of the 2a3 sound, my favourite being the valve 171a. Anyways, there was a lot of added noise and distortion. It sounds like the F4 will do much better with other tube amplifiers like the 300b, but the RCA 45 (1.5W) couldn't provide enough voltage before distorting. So I learnt something and stopped trying to use my tube amplifier.
So back to simplicity, my phono preamp seems to have enough voltage to drive the F4 (it has a TVC volume control) and the DAC too. The dac has a class A output stage that reaches an output up to 10V as single-ended and heats to 50C, I like that.
Today I was revisiting the setup, checking bias and gain, and particularly the Left channel seems mismatched. 0.47ohm3W resistors can read quite differently, so today I just adjusted them so they average 200mV bias. The transistors are from the store and the amplifier is the original design, I only changed R9 as per the instructions to be able to increase bias. I am attaching an image with the measurements.
The power supply voltage is +21.7V, 0V, -21.7V, I got the transformer from Europe so I suspect that is why I get a slightly lower voltage compared with the 23V recommended in the design.
LEFT CHANNEL
Rails 21.7V
offset 0V
R21 169.5mB
R20 244mV
R19 191mV
R18 206mV
R17 198mV
R16 199mV
RIGHT CHANNEL
RAILS 21.7V
OFFSET 0V
R21 217mV
R20 188mV
R19 199mV
R18 216mV
R17 191mV
R16 197mV
R21 and R20 at the left channel are suspects of mismatch...
QUESTIONS:
- How bad is the bias mismatch situation? I read it can be risky, but is it also affecting sound quality? I have another bag with F4 transistors, should I just replace them and see? Can I replace only R21, R20, R19 in one channel or do I have to replace everything? I don't understand the circuit that well, to be honest, and what are the consequences of the mismatch
- Could I benefit from implementing a PRE BA3 stage for the F4? Is it possible to get better sound quality? If recommended, should I implement BA3 separately or build a whole new amplifier with the BA3 pre and power boards?
- I've reached a point where speakers and sources are kind of fancy (for this simple peasant) so any recommendation to improve SQ is welcomed, so far I am quite happy with the results, thank you, everybody, for the generosity and for having this great community.
Carlos
BC, Canada.
It took me a few months to report back
I got busy getting a DAC and building a streamer, just tired of records being so expensive, plus lack of space plus wanting a good-sounding option for exploring new music...The F4 works and it's great! I tested it with my 94db speakers and my ultra-low power SET and realized that the SET simply doesn't have it. enough voltage gain. It works better with the 2a3 triode, but I never liked the sound of the 2a3 sound, my favourite being the valve 171a. Anyways, there was a lot of added noise and distortion. It sounds like the F4 will do much better with other tube amplifiers like the 300b, but the RCA 45 (1.5W) couldn't provide enough voltage before distorting. So I learnt something and stopped trying to use my tube amplifier.
So back to simplicity, my phono preamp seems to have enough voltage to drive the F4 (it has a TVC volume control) and the DAC too. The dac has a class A output stage that reaches an output up to 10V as single-ended and heats to 50C, I like that
.Today I was revisiting the setup, checking bias and gain, and particularly the Left channel seems mismatched. 0.47ohm3W resistors can read quite differently, so today I just adjusted them so they average 200mV bias. The transistors are from the store and the amplifier is the original design, I only changed R9 as per the instructions to be able to increase bias. I am attaching an image with the measurements.
The power supply voltage is +21.7V, 0V, -21.7V, I got the transformer from Europe so I suspect that is why I get a slightly lower voltage compared with the 23V recommended in the design.
LEFT CHANNEL
Rails 21.7V
offset 0V
R21 169.5mB
R20 244mV
R19 191mV
R18 206mV
R17 198mV
R16 199mV
RIGHT CHANNEL
RAILS 21.7V
OFFSET 0V
R21 217mV
R20 188mV
R19 199mV
R18 216mV
R17 191mV
R16 197mV
R21 and R20 at the left channel are suspects of mismatch...
QUESTIONS:
- How bad is the bias mismatch situation? I read it can be risky, but is it also affecting sound quality? I have another bag with F4 transistors, should I just replace them and see? Can I replace only R21, R20, R19 in one channel or do I have to replace everything? I don't understand the circuit that well, to be honest, and what are the consequences of the mismatch
- Could I benefit from implementing a PRE BA3 stage for the F4? Is it possible to get better sound quality? If recommended, should I implement BA3 separately or build a whole new amplifier with the BA3 pre and power boards?
- I've reached a point where speakers and sources are kind of fancy (for this simple peasant) so any recommendation to improve SQ is welcomed, so far I am quite happy with the results, thank you, everybody, for the generosity and for having this great community.
Carlos
BC, Canada.
Attachments
if you have trustworthy set of matched mosfet triplets, mount those instead of existing ones
yup, BA3 FE should certainly drive F4 properly to full power
are those L300 JBLs? if yes, Pa posted complete article how to make them better: https://www.firstwatt.com/pdf/art_l300.pdf
yup, BA3 FE should certainly drive F4 properly to full power
are those L300 JBLs? if yes, Pa posted complete article how to make them better: https://www.firstwatt.com/pdf/art_l300.pdf
Ok so to have things matched I only need to change the 3 P-channel Mosfets on that channel (Q6,Q7,Q10 in the first post diagram)?
I have another F4 kit bag and I can try my luck.
1) what are the effects of mismatching? will mismatch cause more DISTORTION?
2) Can I verify the matching with a multimeter or it's necessary to have a power supply to do proper matching? They are meant for the kit so chances are I will get better results but the bag has 6 p-ch and 6 n-ch , so ideally there is a way to pick the better ones... how does that work?
Yup, those are L300, I've been thinking about implementing Papa's crossover, mostly because these are old speakers, I suspect they would benefit from new capacitors. Just not sure if I benefit more from that first, versus implementing a BA3 preamp first. My dac is not euphonic at all, and I wonder If I would benefit from tweaking harmonic content and sound signature with the BA3... If someone has an opinion please let me know.
best,
C
I have another F4 kit bag and I can try my luck.
1) what are the effects of mismatching? will mismatch cause more DISTORTION?
2) Can I verify the matching with a multimeter or it's necessary to have a power supply to do proper matching? They are meant for the kit so chances are I will get better results but the bag has 6 p-ch and 6 n-ch , so ideally there is a way to pick the better ones... how does that work?
Yup, those are L300, I've been thinking about implementing Papa's crossover, mostly because these are old speakers, I suspect they would benefit from new capacitors. Just not sure if I benefit more from that first, versus implementing a BA3 preamp first. My dac is not euphonic at all, and I wonder If I would benefit from tweaking harmonic content and sound signature with the BA3... If someone has an opinion please let me know.
best,
C
https://www.firstwatt.com/pdf/art_mos_test.pdf
Some questions replied already 20 years ago, just for this very moment
Some questions replied already 20 years ago, just for this very moment
just change that complete triplet where L channel R21 is
mismatch is not going to give much distortion in lower signal range but can cause some with higher power
though, it is just bad practice, in some border cases even leading to malfunction
- if you want better speakers, just do it per Pa's article
- BA3 FE could give you some flavor of 2nd dominant THD, simply said - you'll have some sugar in sound; now - considering sick output level of your DAC, it is necessary to set gain of BA3 FE to accommodate that
now, it is matter of few things - what you use as volume setting device, in which way you have solved impedances in signal chain (pre and post volume) and do you have enough SPL, as it is now
in short - do you have anything missing in system functionality, and do you have anything missing sound-wise
mismatch is not going to give much distortion in lower signal range but can cause some with higher power
though, it is just bad practice, in some border cases even leading to malfunction
- if you want better speakers, just do it per Pa's article
- BA3 FE could give you some flavor of 2nd dominant THD, simply said - you'll have some sugar in sound; now - considering sick output level of your DAC, it is necessary to set gain of BA3 FE to accommodate that
now, it is matter of few things - what you use as volume setting device, in which way you have solved impedances in signal chain (pre and post volume) and do you have enough SPL, as it is now
in short - do you have anything missing in system functionality, and do you have anything missing sound-wise
Thank you ZenMod,
I will start changing the triplet and will report back!
I've been avoiding work but probably I should setup a soundcard distortion analyzer to know better what magic is going on with those matched paralleled MOSFETS.
Regarding the DAC, it has digital volume control and trimmers to adjust the gain so it could adjust to any amplifier, it's just me wondering how to improve SQ in the chain even further and where is the easy cheese.... so far the turntable is still sounding better than the DAC
The Diyaudio Store claims to have matched the MOSFETS at 23V and 0.4A. but the source current for each MOSFET should be (v=ri) 200mV/0.47ohms = 0.43A. If we were to assume a good matching the sweet spot should be at .47ohms*0.4A = 188ohms, despite the claim that distortion should go down at a higher bias?! Other people claim 300mV of bias is better, too many variables... I am building that distortion analyzer cable for the soundcard
Cheers!
I will start changing the triplet and will report back!
I've been avoiding work
but probably I should setup a soundcard distortion analyzer to know better what magic is going on with those matched paralleled MOSFETS.Regarding the DAC, it has digital volume control and trimmers to adjust the gain so it could adjust to any amplifier, it's just me wondering how to improve SQ in the chain even further and where is the easy cheese.... so far the turntable is still sounding better than the DAC
The Diyaudio Store claims to have matched the MOSFETS at 23V and 0.4A. but the source current for each MOSFET should be (v=ri) 200mV/0.47ohms = 0.43A. If we were to assume a good matching the sweet spot should be at .47ohms*0.4A = 188ohms, despite the claim that distortion should go down at a higher bias?! Other people claim 300mV of bias is better, too many variables... I am building that distortion analyzer cable for the soundcard
Cheers!
I'm running two borrowed F4 monoblocks balanced and really like them, but I don't like two 4U cases sitting around. Has anyone tried putting two F4s in a deep (500mm) 5U case? The two F4s (in 400mm cases) don't get very warm so I thought it might be possible to put two in one case if I push the temp a little bit.
I haven't done that, but I've put a 6-deep BA-3 (6 output pairs) in a 4U/500. The temp coefficient for a 4U/500 and a 5U/400 should not be that meaningfully different... Modushop has data posted for the 5U/400 sinks, but I'm not sure about the individual sinks for the 4U/500. Stacking the boards (the only way I can think to get them both on one pair of sinks in a 400mm deep chassis) may be less than ideal, but I don't know how you plan to mount the boards or if you're using "store boards".
Either way... anecdotally... it's hot, but it's fine. Others may differ in their thoughts.
Edited to fix two typos... and to add the thought that I'm assuming that you're comparing apples to apples with Modushop sinks vs. some other manufacturer's.
Either way... anecdotally... it's hot, but it's fine. Others may differ in their thoughts.
Edited to fix two typos... and to add the thought that I'm assuming that you're comparing apples to apples with Modushop sinks vs. some other manufacturer's.