chrisb / bm2352 - I'm also considering combining two ACA's into one chassis. Interesting point about the switching power supply. Are you suggesting placing it inside the amp chassis? I'd always understood switching supplies to be noisy. Is this one quiet enough that it's not an issue ? Did Nelson Pass spec the outboard supply for the ACA purely for cost reasons or was noise reduction part of the consideration?
1) Sure! Go ahead if you want. (The bricks are really good though...)
2) Yes, everybody that has built the kit. If you want to spend $11 a resistor on bulk film or something like that, go ahead. Don't think it's going to make a huge difference in sound quality however. Mainly it just makes your wallet lighter.
3) Again, if you are interested in that, go ahead! There is quite a bit of excess room and there are a ton of caps to try.
4) Yes. Build it as designed first and make changes one at a time so you understand what things actually do to the sound. That way you'll appreciate just how good a deal it is in it's stock form.
Thanx for all the responses, yes I hear you, I’ve modded/ built literally hundreds of diy projects here, a lot of them in pairs as microphones and preamps which kinda call for pairs, especially console channel strips x 50, each channel has mic pre, hp/Lp filter stages, multiple eq stages, signal routing circuits etc.... I’ve taken these opportunities to build them with multiple different component choices, I have heard the subtle and not so subtle differences these mods reveal with many blind ABX listening tests with myself and pals, nothing ever seems to pack a punch like a serious PSU upgrade IME. I was curious to see what people have tried and thought with these little guys as I’ve never built single ended class a power amps, I’m looking forward to it!
Thanx again
Krazy - firstly, be aware that the new ACA1.6 as Jason is now referring to it, is a dual channel amp in a very nice looking 2U chassis. They can be run in bridged mode (with balanced input signal) to achieve higher power.
What bm2352 is referring to is building 4 channels - i.e. a pair of stereo amps - to allow for that option - in a single chassis. As the ACA is a ClassA SE amp, they can run kinda warm, and the concern I expressed was regarding dissipating double the heat in a single 3U chassis of the same depth. I think a 4U would be better suited, but could be wrong.
The latest generation of SMPS - particularly the laptop brick types - are mechanically and electrically quiet, and I think that in keeping with the budget target in mind, Nelson & Jason's choice of the 19V and 24V supplies was mostly one of cost and simplicity. This type of supply is built in huge quantities, and the lower wattage 19V model likely costs less than the mains transformer alone required for a linear supply.
My suggestion of placing a caged supply something like this internally was due as much as anything to my personal preference for as few boxes, dongles, interconnects, etc as possible. I've built a couple of kits and bought cheap classD amps using internal switching supplies, and they were clean enough for my use, but as always YMMV.
What bm2352 is referring to is building 4 channels - i.e. a pair of stereo amps - to allow for that option - in a single chassis. As the ACA is a ClassA SE amp, they can run kinda warm, and the concern I expressed was regarding dissipating double the heat in a single 3U chassis of the same depth. I think a 4U would be better suited, but could be wrong.
The latest generation of SMPS - particularly the laptop brick types - are mechanically and electrically quiet, and I think that in keeping with the budget target in mind, Nelson & Jason's choice of the 19V and 24V supplies was mostly one of cost and simplicity. This type of supply is built in huge quantities, and the lower wattage 19V model likely costs less than the mains transformer alone required for a linear supply.
My suggestion of placing a caged supply something like this internally was due as much as anything to my personal preference for as few boxes, dongles, interconnects, etc as possible. I've built a couple of kits and bought cheap classD amps using internal switching supplies, and they were clean enough for my use, but as always YMMV.
An externally hosted image should be here but it was not working when we last tested it.
Slam dunk.
Order in for two units.
The plan is to go for balanced monoblocks.
Geez, what have I done? I'm going to be in trouble once these dudes arrive, the joy of juggling time to get the build finished. Something tells me the end result will be motivation enough.
Thanks to Nelson and diyA for making this possible.
Order in for two units.
The plan is to go for balanced monoblocks.
Geez, what have I done? I'm going to be in trouble once these dudes arrive, the joy of juggling time to get the build finished. Something tells me the end result will be motivation enough.
Thanks to Nelson and diyA for making this possible.
Chrisb, thanks for the input. In your estimation, would that smps you recommended be enough to power the whole thing (ie, all four)? Also, just to be clear, because I'm new to this and perhaps not using the correct terminology, the plan I'm trying to describe is basically putting a pair of bridged monoblocks together in one chassis to operate as one 15w stereo amp (you seemed to get the gist but I've confused some other people trying to describe it). I'm happy to use two separate power supplies, as proscribed, though your idea of putting one power supply inside the unit is a neat one.
This is a sad thing really. My quote for a single KIT to Ecuador, South America is $121.63 for the cheap option and $140 for the express one.
It really makes little sense to pay $317 for parts and $140 for shipment. That is before customs, taxes and what fedex charges locally for 'storage' (which nobody asked from them, obviously). So I may be looking at around a 100% markup over the selling price.
It is the way of the world: average household salary in the US is around $5000 a month, here, its $2000 for a good earning family. So while in the US this would be a 6% of the monthly salary, for me it would be 30% of mine.
This kind of takes much of the fun away. Also, it's what keeps third world countries actually third world countries. If rich countries get stuff cheaper than poorer countries, then nothing will change.
I don't want to turn this into politics. I really thank Jason and others all the effort to bring business to international customers, but I would rather pay you more money than spend half the budget in shipment.
I am just venting my frustration, as there is little to be done.
Thanks again,
Rafa.
Last edited:
Rafa ....... considering how ACA is simple , is there posibility to make it from parts sourced locally ?
hey , scrapyards are perfect for obtaining parts for funny cases
then PSU can be made in myriad ways , all of them cheap
mosfets - practically anything can work here , with small sonic compromise or even without
hey , scrapyards are perfect for obtaining parts for funny cases
then PSU can be made in myriad ways , all of them cheap
mosfets - practically anything can work here , with small sonic compromise or even without
Exactly this.
Where there's a will there's a way.
This mad simple circuit should easily be parts sourced locally, and wired point to point.
Hi Rafa,
You've also posted here there may be some concern about the suitability of the ACA
with your Totem speakers:
http://www.diyaudio.com/forums/pass-labs/320867-aca-v1-1-illustrated-build-guide-31.html#post5430644
Given that, I would actually suggest you try a cheap & cheerful build with local parts
and keeping the cost down. If it's not to your liking, no big loss and you at least
had some fun and probably learned something.
( BTW, shipping to Ecuador by courier seems really expensive. For example, I looked
on Digikey (where I sometimes buy parts) and they charge a rate of $100USD for
shipping: https://www.digikey.com/en/help/set-rate-shipping/ec )
Dennis
You've also posted here there may be some concern about the suitability of the ACA
with your Totem speakers:
http://www.diyaudio.com/forums/pass-labs/320867-aca-v1-1-illustrated-build-guide-31.html#post5430644
Given that, I would actually suggest you try a cheap & cheerful build with local parts
and keeping the cost down. If it's not to your liking, no big loss and you at least
had some fun and probably learned something.
( BTW, shipping to Ecuador by courier seems really expensive. For example, I looked
on Digikey (where I sometimes buy parts) and they charge a rate of $100USD for
shipping: https://www.digikey.com/en/help/set-rate-shipping/ec )
Dennis
Hi to ACA community.This is my ACA made from materials found mostly at home.For SK170 not shure but IRFP250 is probably fakes PCB is made at home by the suggestion of Rudi Ratlos.Sounds good to me and I'll made another one with proper elements.
Nuki
Nuki
Attachments
Just ordered the components pack without power supply. Shipping to The Netherlands was 30 dollars. Including the Italien housing and tax I expect costs will be around 300 Euro. Will use an old laptop power supply at first.
I now have two ACA mono blocks ready for test.
Will start the job making the linear 24V PSUs for them. I also have a lab supply comming soon. Then I can test with current limiter etc.
I used PCB screw terminals for easy maintenance for this proto type like version of the ACA. 2 x 3300 uF Jensen audio grade output capacitors and a KEMET R75 MKP 10 uF / 90VAC as input capacitor. It has very good high frequency performance according to data and measurements. Will see how it sounds.....
For the two parallel Jensen caps. I have following measurements:
10 Hz, Z = 1.8 Ohm
100 Hz, Z = 193 mOhm
1 kHz, Z = 22 mOhm
5 kHz, Z = 8.8 mOhm
10 kHz, Z = 7.9 mOhm
20 kHz, Z = 7.5 mOhm
50 kHz, Z = 8.1 mOhm
100 kHz, Z = 9.1 mOhm
It has its lowest impedance around 20 kHz but it still very low at 100 kHz indicating low inductance. The theoretical impedance 1/(2 * Pi * f * C) is impossible. So now I know why large capacitors switch from negative to positive phase angle at high frequencies. Few weeks ago I considered that a bad sign for a capacitor but now I can see why.....
I made a few measurements on the output cap in the kit:
10 Hz, Z = 5.5 Ohm
1 kHz, Z = 82 mOhm
100 kHz, Z = 61 mOhm
Interesting that the output level is only about 0.5 dB down at 10 Hz in 8 ohm when the impedance of the capacitor is relative high at 10 Hz. It must be the NFB that lowers the ouput impedance.
Maybe my setup can output more power at the lower frequencies......will see.....first it needs to be tested at see if it work at all.....but so far so good......
Will start the job making the linear 24V PSUs for them. I also have a lab supply comming soon. Then I can test with current limiter etc.
I used PCB screw terminals for easy maintenance for this proto type like version of the ACA. 2 x 3300 uF Jensen audio grade output capacitors and a KEMET R75 MKP 10 uF / 90VAC as input capacitor. It has very good high frequency performance according to data and measurements. Will see how it sounds.....
For the two parallel Jensen caps. I have following measurements:
10 Hz, Z = 1.8 Ohm
100 Hz, Z = 193 mOhm
1 kHz, Z = 22 mOhm
5 kHz, Z = 8.8 mOhm
10 kHz, Z = 7.9 mOhm
20 kHz, Z = 7.5 mOhm
50 kHz, Z = 8.1 mOhm
100 kHz, Z = 9.1 mOhm
It has its lowest impedance around 20 kHz but it still very low at 100 kHz indicating low inductance. The theoretical impedance 1/(2 * Pi * f * C) is impossible. So now I know why large capacitors switch from negative to positive phase angle at high frequencies. Few weeks ago I considered that a bad sign for a capacitor but now I can see why.....
I made a few measurements on the output cap in the kit:
10 Hz, Z = 5.5 Ohm
1 kHz, Z = 82 mOhm
100 kHz, Z = 61 mOhm
Interesting that the output level is only about 0.5 dB down at 10 Hz in 8 ohm when the impedance of the capacitor is relative high at 10 Hz. It must be the NFB that lowers the ouput impedance.
Maybe my setup can output more power at the lower frequencies......will see.....first it needs to be tested at see if it work at all.....but so far so good......
Attachments
Hi all...figured I’d post a quick pic of what I’m doing with the ACA 1b boards and chassis I picked up a month or so back in the swap meet.
Since they were so cheap and simple I felt compelled to try building them before moving on from my gainclone 3875 that I built a few years back into an F2J project. I sourced the components with Nelson’s original BOM. Made a few substitutions for NLA items and additional resistors needed for subsequent modifications to the circuit. Ordered it at the same time I ordered parts for the F2J from Mouser and Digi Key.
I like to build from scratch and make my projects unique in some way, but this time around having the ACA cases made me think about other options.
I thought I was going to build big linear power supplies in separate cases that could also be used with the F2J, but I wanted to go A to B more with the ACA so I could have a new toy to listen to.
I did take time to set a way up to use a few unmatched backup Semisouth R100 I have in the ACAs. I didn’t know if I wanted them to stay in the ACA, so I made some sockets with extension harness’ for them to connnect to the PCB which I relocated to the floor of the case.
I had some 19V laptop supplies to try, but decided to just pony up for the new 24V Meanwell supplies and “cut to the chase” so to speak.
I also did a little rearranging on the back panel as the 5 way posts I had on hand fit directly into the horizontal pair of holes and seemed to make routing wires a little better.
Here’s the first one together and fired up for an initial test and bias adjustment. The only test equipment I have is my ears. Not sure how things will go or if I’ll be compelled to swap in different transistors, but the option is there.
Since they were so cheap and simple I felt compelled to try building them before moving on from my gainclone 3875 that I built a few years back into an F2J project. I sourced the components with Nelson’s original BOM. Made a few substitutions for NLA items and additional resistors needed for subsequent modifications to the circuit. Ordered it at the same time I ordered parts for the F2J from Mouser and Digi Key.
I like to build from scratch and make my projects unique in some way, but this time around having the ACA cases made me think about other options.
I thought I was going to build big linear power supplies in separate cases that could also be used with the F2J, but I wanted to go A to B more with the ACA so I could have a new toy to listen to.
I did take time to set a way up to use a few unmatched backup Semisouth R100 I have in the ACAs. I didn’t know if I wanted them to stay in the ACA, so I made some sockets with extension harness’ for them to connnect to the PCB which I relocated to the floor of the case.
I had some 19V laptop supplies to try, but decided to just pony up for the new 24V Meanwell supplies and “cut to the chase” so to speak.
I also did a little rearranging on the back panel as the 5 way posts I had on hand fit directly into the horizontal pair of holes and seemed to make routing wires a little better.
Here’s the first one together and fired up for an initial test and bias adjustment. The only test equipment I have is my ears. Not sure how things will go or if I’ll be compelled to swap in different transistors, but the option is there.
Attachments
