Upgrade update.
My step 1 upgrades to my amps included the Tungsten mods including IRFP140 transistors, I found the results of these to be fantastic and used the amps in parallel mono.
I just finished step 2 on one of my amps, which included replacing the remaining resistors with Vishay RN series, replacing Q1 with 512-FQH44N10-F133 with hopes to flesh out the midrange and further improve bass. I added 1000uv 50v Nichicon KL's on the DC input and followed Craigl59's advice in repurposing the rear switch as a power switch and running two meanwell smps, on for each board.
The results right off the bat are great, the highs are smoother and more detailed. The vocals are much richer, the bass is much deeper and more articulated. The overall effect is that everything is more tonally correct. I am really surprised.
My next step is to upgrade the next amp and order two more meanwell supplies to run Parallel mono blocks with a total of four power supplies.
Does anyone think there is a potential issue running the amps as parallel monoblocks with the double smps setup?
Thanks to everyone for all the help and the amazing suggestions.
My step 1 upgrades to my amps included the Tungsten mods including IRFP140 transistors, I found the results of these to be fantastic and used the amps in parallel mono.
I just finished step 2 on one of my amps, which included replacing the remaining resistors with Vishay RN series, replacing Q1 with 512-FQH44N10-F133 with hopes to flesh out the midrange and further improve bass. I added 1000uv 50v Nichicon KL's on the DC input and followed Craigl59's advice in repurposing the rear switch as a power switch and running two meanwell smps, on for each board.
The results right off the bat are great, the highs are smoother and more detailed. The vocals are much richer, the bass is much deeper and more articulated. The overall effect is that everything is more tonally correct. I am really surprised.
My next step is to upgrade the next amp and order two more meanwell supplies to run Parallel mono blocks with a total of four power supplies.
Does anyone think there is a potential issue running the amps as parallel monoblocks with the double smps setup?
Thanks to everyone for all the help and the amazing suggestions.
Zamjazz27: Cannot comment on the parallel monoblock approach but am much interested in your results. Am finding, after a full day of listening, that the 2-smps approach has enough power for my system but I am using Tekton Double Impacts and they are known for handling flea-watt amps with grace and power.
Have found two particular listening qualities with my build: 1) loud passages, and especially dramatic ones such as tympani strikes, are realistic, jaw-dropping in power, effortless, and tonally independent -- meaning that the other voices remain at their real world level. SO the ACA is not nearly as loud overall as my other amps but GETS LOUD BETTER.
2) The soundstage is incredibly good with a great deal of space between voices and 3D depth keeping all in the right position. Would like your assessment here as I speculate that some of this is the result of using a single circuit/smps for each channel and not confusing the signal by combining a single channel with 2 or more circuits -- as you will be doing with the parallel approach. So if your soundstage is also like the above, then I will know that multiple circuits can be combined without detriment and can start ordering additional ACAs.
Again, I do not need more power because of my setup. Have been listening all day at much higher dB levels than normal -- measuring 85-95dB at the listening chair. The musical texture is open and clean with a lot of space so that louder settings are more natural and appealing.
This is one fine amp and the improvements made by this thread show the internet in its best light of offering information with accuracy and custom specifics.
Have found two particular listening qualities with my build: 1) loud passages, and especially dramatic ones such as tympani strikes, are realistic, jaw-dropping in power, effortless, and tonally independent -- meaning that the other voices remain at their real world level. SO the ACA is not nearly as loud overall as my other amps but GETS LOUD BETTER.
2) The soundstage is incredibly good with a great deal of space between voices and 3D depth keeping all in the right position. Would like your assessment here as I speculate that some of this is the result of using a single circuit/smps for each channel and not confusing the signal by combining a single channel with 2 or more circuits -- as you will be doing with the parallel approach. So if your soundstage is also like the above, then I will know that multiple circuits can be combined without detriment and can start ordering additional ACAs.
Again, I do not need more power because of my setup. Have been listening all day at much higher dB levels than normal -- measuring 85-95dB at the listening chair. The musical texture is open and clean with a lot of space so that louder settings are more natural and appealing.
This is one fine amp and the improvements made by this thread show the internet in its best light of offering information with accuracy and custom specifics.
Craigl59, that is really interesting as I found I was listening louder than usual as well, I think the benefit is that the power is so clean, I don't know how to describe it. There is nothing that is fatiguing yet it is very detailed. YES! things lie timpani strikes are jaw-dropping! I know you must be used to awesome bass from your double impacts, but the bass coming out of the bookshelf Klipsch rp600m with the amp setup this way seems to defy the laws of time and space...
I am curious to hear how the parallel monoblocks will sound, because with just the Tungsten mods going to monoblocks made a huge difference, and I expect into your 4-ohm speakers it would be great.
On a side note, I have a good friend trying to choose between the double impacts and the moabs. Great to hear that the Tektons sound great with this amp...I would love to get some in the future.
I am curious to hear how the parallel monoblocks will sound, because with just the Tungsten mods going to monoblocks made a huge difference, and I expect into your 4-ohm speakers it would be great.
On a side note, I have a good friend trying to choose between the double impacts and the moabs. Great to hear that the Tektons sound great with this amp...I would love to get some in the future.
Zamjazz27: Have two pair of the Tekton DIs and have settled on them for these reasons: The drivers they have are professional audio ones and you can order spares and replacements yourself at modest cost -- I have complete sets for my speakers so no blow out would be a problem. Second, the 11 speakers in each cabinet move a LOT of air. So if you want to recreate orchestral sounds, you've got this need in spades. Third, the overall FR is superb as long as you damp the back port -- this is critical. Don't let it wash into the wall, reverberate, then spit back out towards the listener. Kill it completely with a baffle made out of foam or whatever.
The total costs for all of the ACA upgrades I am using at Mouser is $50 + postage. Am preparing another order now. Consequently, this means the total 2-smps ACA with upgrades cost with shipping is around $460. I haven't heard any other amp so far that competes at the flea level with this 2-smps build (but there are some, I suspect).
Wonder how the 25-watt Pass integrated amps compare and am worried that my ACA experience here might make it necessary to buy one or more...
The total costs for all of the ACA upgrades I am using at Mouser is $50 + postage. Am preparing another order now. Consequently, this means the total 2-smps ACA with upgrades cost with shipping is around $460. I haven't heard any other amp so far that competes at the flea level with this 2-smps build (but there are some, I suspect).
Wonder how the 25-watt Pass integrated amps compare and am worried that my ACA experience here might make it necessary to buy one or more...
Thanks Craigl59!
I've just been doing some A/B listening between my "Tungsten mods" ACA and my double smps/filter cap/all vishay/Q1=FQH44N amp
They both sound really good, and I did a few mods all at once so it can be challenging to say what change caused what, but I would think that the switch to all Vishay resistors would be subtle?
So after the A/B listening:
The dual smps amp has bass that I would equate to the bass I had from the amps in parallel mono, at least as good. It has a more focused and deeper sound. The bass notes are more centered on the precise note they are playing. Mids are richer and fuller, vocals sound really natural (this may also be the FHQ44N at Q1? I still have a 140 at Q2. The depth and impact of the bass and bass transients are really shocking. The definition of the attack on an acoustic bass is fantastic.
The upper mids and highs are much more natural, the different touch on the cymbals, brushes on a snare all sound eerily real.
I think Craigl59 was so on the money by saying it plays louder better, not something I'd expect from an 8-watt amp.
It would be interesting to compare it in this iteration to a Schitt Aegir?
I'm getting to work on the second amp and will report back on parallel mono with a total of four smps!
I've just been doing some A/B listening between my "Tungsten mods" ACA and my double smps/filter cap/all vishay/Q1=FQH44N amp
They both sound really good, and I did a few mods all at once so it can be challenging to say what change caused what, but I would think that the switch to all Vishay resistors would be subtle?
So after the A/B listening:
The dual smps amp has bass that I would equate to the bass I had from the amps in parallel mono, at least as good. It has a more focused and deeper sound. The bass notes are more centered on the precise note they are playing. Mids are richer and fuller, vocals sound really natural (this may also be the FHQ44N at Q1? I still have a 140 at Q2. The depth and impact of the bass and bass transients are really shocking. The definition of the attack on an acoustic bass is fantastic.
The upper mids and highs are much more natural, the different touch on the cymbals, brushes on a snare all sound eerily real.
I think Craigl59 was so on the money by saying it plays louder better, not something I'd expect from an 8-watt amp.
It would be interesting to compare it in this iteration to a Schitt Aegir?
I'm getting to work on the second amp and will report back on parallel mono with a total of four smps!
Zamjazz27: Your listening comments are perceptive and very helpful to me. Have just ordered another ACA with an additional smps and all of the upgrades used on the first build -- that include the better 4 Mosfets. Will be following your comparison of mono versus stereo to see if it is worth rewiring mine for mono block formatting. Need another ACA for my winter place and so would like to use just a single, stereo unit if it sounds as good as this first one does.
NOW for some synergy news. Find that the signal source favors certain amp types. My cassette decks really shine with tube amplification, digital sources pair nicely with the powerful Emotiva XPA-1, AND now I have found that vinyl feeds are perfect for the multi-amps ACA. Records are SO MUCH better through this amp that I have been wondering why.
Use Dr. John recordings to check for vocal clarity and the ACA pulls his voice out front and center and makes his most complicated statements clear and alive. Again, there is so much space in the soundstage, and voice are so tightly defined, that the text can be understood better than with either of my other 2 amp types. This synergy is SO GOOD that I will be using the ACA exclusively for vinyl feeds. Orchestral recordings also shine with the vinyl/ACA pairing. Cassettes? Not so much here. Will stick with tube amplification.
Am going to do some timbral checks next with studio recordings of my piano and this will tell me how the ACA does in rendering subtle timbre challenges.
NOW for some synergy news. Find that the signal source favors certain amp types. My cassette decks really shine with tube amplification, digital sources pair nicely with the powerful Emotiva XPA-1, AND now I have found that vinyl feeds are perfect for the multi-amps ACA. Records are SO MUCH better through this amp that I have been wondering why.
Use Dr. John recordings to check for vocal clarity and the ACA pulls his voice out front and center and makes his most complicated statements clear and alive. Again, there is so much space in the soundstage, and voice are so tightly defined, that the text can be understood better than with either of my other 2 amp types. This synergy is SO GOOD that I will be using the ACA exclusively for vinyl feeds. Orchestral recordings also shine with the vinyl/ACA pairing. Cassettes? Not so much here. Will stick with tube amplification.
Am going to do some timbral checks next with studio recordings of my piano and this will tell me how the ACA does in rendering subtle timbre challenges.
.. The magic of a simple Class A circuit with the right kind of power supply decoupling. I believe the FQH44N10 parts are helping here as well.
The Sony VFET amps (lottery) have a similar, or even more pronounced ease of dynamic presentation. Beautiful subtlety in quiet passages and nearly startling impact with percussion instruments. Magic with bass guitar. They have responded well to the same kind of PSU treatment. Consider also that I eventually built a full linear, dual-mono, PSU for my ACA and VFET amps. Makes them reach even deeper into the lower octaves, allowing the circuits to sound their best.
The Sony VFET amps (lottery) have a similar, or even more pronounced ease of dynamic presentation. Beautiful subtlety in quiet passages and nearly startling impact with percussion instruments. Magic with bass guitar. They have responded well to the same kind of PSU treatment. Consider also that I eventually built a full linear, dual-mono, PSU for my ACA and VFET amps. Makes them reach even deeper into the lower octaves, allowing the circuits to sound their best.
TA: Thanks for this info. Noted that a 39V Meanwell smps was available at DIY and was wondering whether the ACA circuit boards could be "upsized" to use this larger PU and with adjusted mosfets (or whatever) to produce greater output watts. Don't have the technical skills, obviously, but wonder what the path towards greater power is should you wish to stay class A with as little circuit complexity as possible.
Am preoccupied with the concept "continuity." Analog methods produce a continuous record of the sound that is immediately apparent with true analog vinyl (and a class a analog phono pre). So this afternoon have been playing my Direct-to-Disc records through the ACA. The waveform is maintained through this entire process and the result is "goose-bump" physicality. Haitink's DtoD recording's of Bruckner's 7th (released by BP last year) was listened to at 95+ dB, higher than I have ever endured with A/B. And the bass was over-the-top good with seamless, powerful crescendos never heard before from this record. This little ACA can be the loudest amp in my collection because the loudest sounds it makes are easy to listen to, natural, and only the music grabs your attention.
Looking forward to the second build and to examining other amp types as my knowledge grows. Again, your help has been greatly appreciated and it was your suggestion that the dual smps approach might work that made me give it a try. Think it has succeeded.
Am preoccupied with the concept "continuity." Analog methods produce a continuous record of the sound that is immediately apparent with true analog vinyl (and a class a analog phono pre). So this afternoon have been playing my Direct-to-Disc records through the ACA. The waveform is maintained through this entire process and the result is "goose-bump" physicality. Haitink's DtoD recording's of Bruckner's 7th (released by BP last year) was listened to at 95+ dB, higher than I have ever endured with A/B. And the bass was over-the-top good with seamless, powerful crescendos never heard before from this record. This little ACA can be the loudest amp in my collection because the loudest sounds it makes are easy to listen to, natural, and only the music grabs your attention.
Looking forward to the second build and to examining other amp types as my knowledge grows. Again, your help has been greatly appreciated and it was your suggestion that the dual smps approach might work that made me give it a try. Think it has succeeded.
To make proper use of that 39v to make full power, you'd need a different design. There's no single mosfet that can do that for you, I think.
Do remember, the ACA and any other class A amp does not at all "retain the waveform". It's literally amongst the worst topologies if that's your goal. The ACA starts to heavily saturate at a pretty low volume. That is by design and preference though - I don't think anyone gets a Class A amp with the impression they are the best amp for transparency! 😛
Regarding using two PSU's in a parallel setup: you will need some way to make sure the voltage differences aren't going to cause issues. That's usually a small resistor on both.
Do remember, the ACA and any other class A amp does not at all "retain the waveform". It's literally amongst the worst topologies if that's your goal. The ACA starts to heavily saturate at a pretty low volume. That is by design and preference though - I don't think anyone gets a Class A amp with the impression they are the best amp for transparency! 😛
Regarding using two PSU's in a parallel setup: you will need some way to make sure the voltage differences aren't going to cause issues. That's usually a small resistor on both.
Last edited:
Tungsten: I got the idea from you for the FQH44N10, I do find that this new setup strikes a great balance, my other amp which still has 140's in both places is much lighter and slightly thinner in the mids, but also can be a little airier. Thanks again for all of the suggestions in this thread! What so think the qualities of that FHQ44N10 that contribute to the sound?
Craigl59: Yes, you can listen loud without any harshness! which mosfets are you going to try?
Humbledeer: Thanks so much for thinking about the possible issues with running parallel mono with the two power supplies. Where would the resistor that you mentioned go? what value?
Any ideas from the hive mind here on how to mitigate any issue that might arise with the double smps in parallel mono.
Thanks!
Craigl59: Yes, you can listen loud without any harshness! which mosfets are you going to try?
Humbledeer: Thanks so much for thinking about the possible issues with running parallel mono with the two power supplies. Where would the resistor that you mentioned go? what value?
Any ideas from the hive mind here on how to mitigate any issue that might arise with the double smps in parallel mono.
Thanks!
Zj: am using the FQH44N10-f133 mosfet, same as the first build. Seems to work fine. Mouser sent info that this part will be "refined" on August 17th with some composition change. Have no idea of the specifics.
The FQH44N10 has a rare combination of high transconductance with a low gate charge and low reverse capacitance. They also have good thermal dissipation. I’m using them in my F6. One thing to note is that their performance improves at higher voltage. I have taken advantage of this.
Given some attention to power dissipation, the ACA could use the same 36V power supply as the VFET lottery amp. R16 needs to be increased to 1k to protect the input JFet. With a 36V rail, the Vds bias setting on Q1 should be about 18.5V.
Given some attention to power dissipation, the ACA could use the same 36V power supply as the VFET lottery amp. R16 needs to be increased to 1k to protect the input JFet. With a 36V rail, the Vds bias setting on Q1 should be about 18.5V.
Thanks, TA. With a Vds setting of 18.5 can you prescribe the watt improvement? As mentioned above, mine is at 12 now but could increase it to 15 with the current setup. Am guessing that this would affect the noise/distortion.
In a sense, my current setup is fine with plenty of volume through the DIs. Just considering future possibilities. Again, can listen to very loud orchestral passages with ease and am astonished at how effortless the dynamic shifts are -- and how clear and "easy listening" loud passages remain. My Emotiva produces a "fuller" sound but it is not so much better as "fatter" in the soundstage arena. This clouds the instruments and obscures some of the 3D placement. ACA excels in positioning inner orchestral forces in their proper stage area L/R and F/B.
In a sense, my current setup is fine with plenty of volume through the DIs. Just considering future possibilities. Again, can listen to very loud orchestral passages with ease and am astonished at how effortless the dynamic shifts are -- and how clear and "easy listening" loud passages remain. My Emotiva produces a "fuller" sound but it is not so much better as "fatter" in the soundstage arena. This clouds the instruments and obscures some of the 3D placement. ACA excels in positioning inner orchestral forces in their proper stage area L/R and F/B.
That's super interesting, would it be beneficial to use the FQH44N10 in both Q1 and Q2? I am only using it in Q1 and have the 140 in Q2.
Also, as the amps ran well in parallel mono with one power supply would there be any reason to think that would change with using two power supplies?
Does the Vfet use the 36v Meanwell? wow, that could be used? super interesting!
Also, as the amps ran well in parallel mono with one power supply would there be any reason to think that would change with using two power supplies?
Does the Vfet use the 36v Meanwell? wow, that could be used? super interesting!
Zj: There is a discussion above concerning the roles of the two mosfets and I would not attempt to describe it but have always used 4 of the same ones and they are not expensive -- about $12 for 4.
At this point we are talking about incremental changes to power output. The parametric changes from running at higher voltage are probably more significant.
We have seen a big improvement in sound quality being reported by using separate power supplies for each channel, as well as adding extra capacitance to the power rails. Increasing the voltage by using 36V Meanwell supplies is just another option that can be explored.
We have seen a big improvement in sound quality being reported by using separate power supplies for each channel, as well as adding extra capacitance to the power rails. Increasing the voltage by using 36V Meanwell supplies is just another option that can be explored.
Thanks TungstenAudio! any thoughts on:
Do you think it is cool to run parallel mono with two power supplies (would there be any harmful interaction?)
What are your thoughts on Q1=FQH44N10 and Q2=IRFP 140 vs going with FQH44N10 for both?
Do you think it is cool to run parallel mono with two power supplies (would there be any harmful interaction?)
What are your thoughts on Q1=FQH44N10 and Q2=IRFP 140 vs going with FQH44N10 for both?
I think the safest configuration is probably a single SMPS per channel, but it looks like we may see a successful build with an SMPS per board (!). So, whatever works. You guys may get to collaborate on a build that mitigates potential difference in oscillator frequencies. Could be that the extra capacitance on the power rails helps with this as well.
I think the choice between using an IRFP140 or FQH44N10 at Q2 is going to be one of personal preference.
I think the choice between using an IRFP140 or FQH44N10 at Q2 is going to be one of personal preference.
So when using one smps per board in the parallel mono configuration, we are not necessarily paralleling the power supplies are we? I wouldn't think that we would need to be concerned with power from one supply backing into the other. The difference in oscillator frequencies is interesting, if that were an issue then I wonder how it would manifest? i have seen a number of good solutions for paralleling smps' but haven't found evidence of this issue being tackled.
I am so tempted to test it out but would hate to lose an amp and two psu's in the process...
I am so tempted to test it out but would hate to lose an amp and two psu's in the process...
Go ahead and try it. There won't be a problem of losing an amp or SMPS, as long as you are reasonably careful with the connections.
The difference in oscillator frequency can potentially manifest itself as a 'beat' frequency that occurs in a non-linear mixing element of the amplifier. (All real electronic devices have some non-linear behavior. Especially silicon) The beat frequency can easily fall into the audio range, even though the primary oscillator is well above audio.
I will leave it to the ACA collective to solve this issue. I know you guys can figure this out. I will be working through other ideas for some time to come. (VFET lottery amp)
The difference in oscillator frequency can potentially manifest itself as a 'beat' frequency that occurs in a non-linear mixing element of the amplifier. (All real electronic devices have some non-linear behavior. Especially silicon) The beat frequency can easily fall into the audio range, even though the primary oscillator is well above audio.
I will leave it to the ACA collective to solve this issue. I know you guys can figure this out. I will be working through other ideas for some time to come. (VFET lottery amp)