The current version (1.8) of the ACA board from the diyAudio store is the one to get. It is convenient for trying different types of stereo vs bridged dual mono operation.
Hi, I just bought a premium ACA off swap meet here that uses these boards. It uses an SMPS, so no linear supply or rectifiers near that ground. I have no hum issue, would it still be wise to relocate the SPK ground? Are the boards otherwise ok to use? Sounds pretty good as it is.
If you have no hum, you probably don't need to make any changes, especially if you have very sensitive speakers. However if your speakers are not very sensitive, then maybe it is worth a further look.
Do you have the cap multiplier parts installed? If you do, then ideally the speaker ground should be moved. If not, then no difference.
Do you have the cap multiplier parts installed? If you do, then ideally the speaker ground should be moved. If not, then no difference.
Thanks, my speakers are not very sensitive. The amp is this one: https://www.diyaudio.com/community/threads/fs-premium-aca-with-cap-mult-eu.384533/
I moved the ground to the PS ground today.
I moved the ground to the PS ground today.
This may be a dumb question but it should at least be easy to answer 😁 After the premium parts upgrade what is the what is the input resistance on the amp. Is it simply going from 10k up to 20k by swapping the input resistor. Or is there more to it than that.
Side note. I upgraded my monoblocks a few months ago. For anyone doing this give things plenty of time to break in. At first they were very bright. I switched to running them in parallel. The way they deliver power to my speakers made this less forward than bridged. I tried bridged again last week and they sound absolutely incredible. The few extra watts delivered into 8 ohms helped some with the headroom and improved the amount of distortion in my system.
I’m looking to add a couple of subs and am considering making an online high pass filter and wanted to get some confirmation in the input impedance with the upgrade so I can try rolling off the low frequency before it hits my amp and have subs blended to take on those duties.
Side note. I upgraded my monoblocks a few months ago. For anyone doing this give things plenty of time to break in. At first they were very bright. I switched to running them in parallel. The way they deliver power to my speakers made this less forward than bridged. I tried bridged again last week and they sound absolutely incredible. The few extra watts delivered into 8 ohms helped some with the headroom and improved the amount of distortion in my system.
I’m looking to add a couple of subs and am considering making an online high pass filter and wanted to get some confirmation in the input impedance with the upgrade so I can try rolling off the low frequency before it hits my amp and have subs blended to take on those duties.
I had to change out some resisitors that I was short in my kit (R11 10K 1ea short I got 10K 1/4 watt)..(R13 10K 2ea short…I got 2ea 10K 1/4 watt)…(R6 100ohm 1 ea short….I got 100 ohm 1/4 watt)…this is what diy audio told me toget…it acts like the wattage is wrong as I have quadruple checked everything….help???? is this the right resistance but the wrong wattage,,,should it have been 3 or 4 watts???? Either way I have no left channel.....Help
Whoa! Are you complaining about getting 1/4 watt resistors or short something in the kit of parts from the store?
The problem is not with the wattage of the resistors.
R11 is the first resistor after the input, 1/8 watt is enough here.
R5/R6 are the gate stoppers, 1/8 watts are enough here too.
R13 is for the LED, this value depends on the LED and the power voltage.
Last edited:
I had the wrong resistor in r15...it was supposed to be 2.21k...so I ordered a new 2.21K 1 watt and took wrong one out...ty 4 ur help
I am thinking of dipping my toes in here and ordering the ACA 1.8 kit, but also buying the recommended premium parts right away and installing them from the get go. After reading through the thread a few times I think I have enough information to pull the trigger on the order(s).
Also a good time to do a shout out to those that have posted tips, updated BoM's, and of course to TungstenAudio for the original recommendations.
My lingering question though is mainly related to how much voltage can I run the Premium ACA rails before adding the R16 mod becomes mandatory. I am not against it in principle or anything, but I just don't like the idea of hacking a new build with a messy trace cut and ugly tack on solder job to accomplish this mod (referring specifically to adding R16).
I plan to power the build with a linear power supply based on a Antek AS4225 (400VA/25V Secondaries). Planning to use each secondary winding to power two rectifier/~60,000 uF capacitor banks to feed each channel separately, and maybe power an eventual dual mono-block setup. I already have the transformer so would like to utilize it.
According to what I have calculated so far, and simulating in LTSpice, I believe I will have roughly ~35V no load present at the capacitor bank with the voltage dropping to around 28 to 28.5V once under load (at the rails). So getting back to the R16 question, is 28 to 28.5V too much without this resistor? I see the two likely trouble spots to be C4 and possibly Q4, which also has been mentioned earlier in this thread.
The other issue would be the momentary turn on voltage of roughly 35V, which would push C4 right to its limit, although I did look at the datasheet for the recommenced upgrade (Nichicon 100uF 35V) and it lists surge voltage as 40.2V.
Any thoughts or reassurances from the trailblazers before me is appreciated.
Also a good time to do a shout out to those that have posted tips, updated BoM's, and of course to TungstenAudio for the original recommendations.
My lingering question though is mainly related to how much voltage can I run the Premium ACA rails before adding the R16 mod becomes mandatory. I am not against it in principle or anything, but I just don't like the idea of hacking a new build with a messy trace cut and ugly tack on solder job to accomplish this mod (referring specifically to adding R16).
I plan to power the build with a linear power supply based on a Antek AS4225 (400VA/25V Secondaries). Planning to use each secondary winding to power two rectifier/~60,000 uF capacitor banks to feed each channel separately, and maybe power an eventual dual mono-block setup. I already have the transformer so would like to utilize it.
According to what I have calculated so far, and simulating in LTSpice, I believe I will have roughly ~35V no load present at the capacitor bank with the voltage dropping to around 28 to 28.5V once under load (at the rails). So getting back to the R16 question, is 28 to 28.5V too much without this resistor? I see the two likely trouble spots to be C4 and possibly Q4, which also has been mentioned earlier in this thread.
The other issue would be the momentary turn on voltage of roughly 35V, which would push C4 right to its limit, although I did look at the datasheet for the recommenced upgrade (Nichicon 100uF 35V) and it lists surge voltage as 40.2V.
Any thoughts or reassurances from the trailblazers before me is appreciated.
- If your wall voltage is a stable 120VAC, unloaded voltage at the secondaries will be a bit higher than 25VAC. Check the Antek specs.
- Are you counting on 6V5 to 7V of sag with 2 channels of ACA and/or are you using a rectification / filtering / regulation set up with that much voltage drop? I wasn't sure if you meant ~35VDC at the unloaded outputs of the full supply. If so, it's doubtful that you'll have that much sag.
Adding R16 does two things – filter the voltage at the drain of the input JFet (Q4) and manage the power dissipation of that device. If you are willing to make the modification to your build, it should help with the sound even where there is no issue with Q4 power dissipation or Vds. The basic value that I use for Q4 is 200Ω.
The next question regards the maximum voltage provided to the amp's power rail based on the secondary voltage spec of the power transformer (when building a linear supply). I recommend a 300VA transformer to keep the size more manageable. Either 24V or 25V secondaries are quite suitable and will deliver plenty of current. The final power rail voltage depends on the type of rectifiers used as well as the supply filter network. You can certainly try CRCRC or even CLCRC filtration. As ItsAllinMyHead suggests, it is unlikely to ever see a full 35V at the output of the filter network. The typical block rectifiers will drop from 2.5V to almost 3.5V depending on their forward voltage drop. I prefer the Vishay VS-26MB05A or VS-26MB10A for lower Vf and ease of installation.
The next question regards the maximum voltage provided to the amp's power rail based on the secondary voltage spec of the power transformer (when building a linear supply). I recommend a 300VA transformer to keep the size more manageable. Either 24V or 25V secondaries are quite suitable and will deliver plenty of current. The final power rail voltage depends on the type of rectifiers used as well as the supply filter network. You can certainly try CRCRC or even CLCRC filtration. As ItsAllinMyHead suggests, it is unlikely to ever see a full 35V at the output of the filter network. The typical block rectifiers will drop from 2.5V to almost 3.5V depending on their forward voltage drop. I prefer the Vishay VS-26MB05A or VS-26MB10A for lower Vf and ease of installation.
Thanks for the replies. I will probably go with something like a Vishay VS-26MB40A for the rectifier as I am following the DIY PSU board build for the initial rectification/filtering. I was going to just use screw terminal caps, but thought that going the PCB route looked a bit cleaner, plus it has space for the resistors.
So I will be doing the Antek AS4225 > Vishay VS-26MB40A > DIY PSU Board with 4x 15,000 uF 50V capacitors in a CRC form. I am indeed getting a stable 120VAC at the wall, so I am confident my unloaded output will be ~34 to 35 VDC.
I guess at this point I can build the PSU first and see as I already have the transformer. I can buy a different 300VA with maybe 22V secondaries, if it comes to it, but I do like the idea of running the ACA a little "hot", meaning maybe 26-27VDC at the rails when in operation.
So all this still leads me back to the the original query of how hot can the stock board configuration take (meaning no R16 mod). I have no doubt in your statement the R16 provides additional benefits, but not willing to make the hack at this point.
I did see you mentioned if someone was up to modifying the board layout in a much earlier post to included spaces for R16 and C101 natively, but didn't see any followup. I know someone else, Rudy I think, had a variation of the board, but it sounded like it included other modifications I would not be interested in.
I might try to take up the challenge myself, as although it may just be my stubberness, a trace cut and solder hack just doesn't sit well when we go to all the trouble of selecting qualtiy components and making a otherwise neat build.
So I will be doing the Antek AS4225 > Vishay VS-26MB40A > DIY PSU Board with 4x 15,000 uF 50V capacitors in a CRC form. I am indeed getting a stable 120VAC at the wall, so I am confident my unloaded output will be ~34 to 35 VDC.
I guess at this point I can build the PSU first and see as I already have the transformer. I can buy a different 300VA with maybe 22V secondaries, if it comes to it, but I do like the idea of running the ACA a little "hot", meaning maybe 26-27VDC at the rails when in operation.
So all this still leads me back to the the original query of how hot can the stock board configuration take (meaning no R16 mod). I have no doubt in your statement the R16 provides additional benefits, but not willing to make the hack at this point.
I did see you mentioned if someone was up to modifying the board layout in a much earlier post to included spaces for R16 and C101 natively, but didn't see any followup. I know someone else, Rudy I think, had a variation of the board, but it sounded like it included other modifications I would not be interested in.
I might try to take up the challenge myself, as although it may just be my stubberness, a trace cut and solder hack just doesn't sit well when we go to all the trouble of selecting qualtiy components and making a otherwise neat build.
I was just thinking that the point that may be adding a bit of confusion of what concern I am describing, is I forgot to mention the PSU will be in a separate chassis from the ACA's, thus my concern with the temporary unloaded voltage when I first power up the ACA.
So I would first turn on the PSU and of course within moments the capacitors charge and are sitting at idle with 34-35VDC at the output. I then go to power on the ACA and it momentarily has that 34-35VDC hit the rails before loading down to the loaded 27VDC or so.
I guess I am not sure how long of a period this is and what, if any, damage to components that could occur as it is right at 35VDC, the C4 capacitors voltage limit, of course taking into consideration it does list a surge voltage of 40.2 V, which I assume is just for such occasions. I would normally just toss a 50V part in that spot, but the recommend Nichicon caps max out at 35V in that series.
So it is basically 2 questions, one dealing with the inital unloaded voltage at the Amps power up (a few milliseconds), and two if the un-modded board (again as in no R16) is capable of handing a constant 27-28V rails during operation.
So I would first turn on the PSU and of course within moments the capacitors charge and are sitting at idle with 34-35VDC at the output. I then go to power on the ACA and it momentarily has that 34-35VDC hit the rails before loading down to the loaded 27VDC or so.
I guess I am not sure how long of a period this is and what, if any, damage to components that could occur as it is right at 35VDC, the C4 capacitors voltage limit, of course taking into consideration it does list a surge voltage of 40.2 V, which I assume is just for such occasions. I would normally just toss a 50V part in that spot, but the recommend Nichicon caps max out at 35V in that series.
So it is basically 2 questions, one dealing with the inital unloaded voltage at the Amps power up (a few milliseconds), and two if the un-modded board (again as in no R16) is capable of handing a constant 27-28V rails during operation.
I wouldn't be to afraid of the trace cut, the boards are so nice and the traces are thick so it is really pretty easy... and you have the benefit of doing it before on empty boards, I did this mod after the fact. Well worth it, I run mine on 36v with FQH44N10's, 1k at R16 and it sounds fantastic.
I went ahead and spent the last 3 days with KiCad (mainly learning and getting used to its quirks) and think I have a board layout that mimics the V1.6 fitment, but also includes spaces for R16, C101 and a few other minor tweaks. I sent the files off to a PCB shop and should have them in a week or two.
I am going to hold off on any further comment on this aspect until I get the boards in-hand and see if they blow up when I populate them and first turn it on. I am buying enough components for a handful of boards just in case. 😀
Now for the harder part, where to find these alternate transistors I see mentioned through-out the thread?
I checked Mouser and Digikey and the FQH44N10's and IRFP140's seem to be out of stock. I even checked Google for some of the lesser known suppliers and not finding much. I would like to avoid eBay due to the greater potential for counterfeits.
I am going to hold off on any further comment on this aspect until I get the boards in-hand and see if they blow up when I populate them and first turn it on. I am buying enough components for a handful of boards just in case. 😀
Now for the harder part, where to find these alternate transistors I see mentioned through-out the thread?
I checked Mouser and Digikey and the FQH44N10's and IRFP140's seem to be out of stock. I even checked Google for some of the lesser known suppliers and not finding much. I would like to avoid eBay due to the greater potential for counterfeits.
I think Arrow Americas has stock, can you check? I looked for FQH44N10 and could find it.
https://www.arrow.com/en/products/fqh44n10-f133/on-semiconductor?q=FQH44N10-F133
Thanks