No noise of any kind, the SMPS is very quiet. So are the ACA's. The ribbons are not enameled, but fairly thick and doubled-up & then twisted for each conductor - so they don't move - no danger of shorting anything. The input signal ribbons are done with 1 silver and 1 copper ribbon for each signal wire (and return). The power and speaker wires are 2 wider copper ribbons, twisted together to form a single conductor.
The current ACA1.6 official diagram calls for 22K feedback resistor, which will give 10dB gain, but also a "solid" damping factor of 10.
There are diagrams circulating around, that have a 68K feedback resistor - this will give a higher gain (13-14dB) but much lower damping factor of only 3. Now, if you use 68K resistor in bridged mode ACA monoblocks, this damping factor will further reduce, which is kind of bad...
I am using a 22K resistor.
Each resistor "pair" (R1&R2, and R3&R4) will dissipate around 0.9 - 1.38W. That is for TWO resistors (in parallel); so each resistor will dissipate around 0.7W in a worst case. They are rated at 3W. I placed them away from the PCB because it just seems right to me to do so...(old habit from the days when I worked with 2000W per channel amplifiers for PA...) The ACA resistors will not burn the PCB even if they are touching the board.
Boyz , I can't cope with amount of your builds .........
so - one collective - Fugly!
Thanks, ZenMod!
Thanks for the reply and I was aware of that issue. However, it wasn’t the problem I had. One of the rails I received was defective and different than the other 3.
If you look at my pics from my first post closely you can see that a gap between the bent portion of the rail and the larger potion of the rail. That gap is not present in the other picture and the gap created my problem. That gap caused the bottom plate to extend below the balance of the case because the screw holes were misaligned until I enlarged them. Check yours for this problem before assembly.
Thanks for the clarification. Maybe you can answer a couple of other questions I have. Does running in balanced bridged mode cause the damping factor to suffer compared to running as a standard stereo amp? Also, does running in balanced bridge mode have ill effects compared to stereo mode if your speakers dip lower than the recommended 8 ohms?
Thanks again.
Aljordan,
Here are a couple of answers (assuming a 24V supply, and assuming the ACA version 1.6 build/schematic):
When running the amp in balanced bridged mode (and if you built it as the 10dB gain option), then the overall gain is 10dB, i.e. the same as stereo.
When running the amp in balanced bridged mode, the output impedance is DOUBLED, i.e 1.6 ohms, thus the damping factor is now 5 (instead of 10, when the output impedance was 0.8 ohms). This is the reason why that in balanced bridged mode, you use speakers that have an 8 ohm nominal impedance. An 8 ohm nominal impedance, assumes that the lowest impedance of that speaker is no less than 80% of that impedance, i.e. about 6.4 ohms or thereabouts (it's not a fixed entity so don't be paranoid if your speaker drops to 6 ohms. But below 6 ohms, I wouldn't recommend it). How do you know the lowest impedance? You measure it or look to an authority that measures it online, i.e. Stereophile and other magazines likes Soundstage, etc...
Running the amplifier in balanced bridged mode is sonically advantageous for a few reasons as long as you use a compatible speaker:
1) In the less than 1 watt area, the overall THD, is very, very low. Even lower than in stereo mode.
2) The overall max output power is increased to about 15 watts-16 watts into 8 ohms and nearly 20 watts into 16 ohms. This is assuming about 3% THD.
Here is the THD vs. output power in balanced bridged mode for an 8 ohm impedance/load:
For a 16 ohm load, the graph is a little better with a little lower distortion:
Will the amplifier explode if you hooked it up to a 4 ohm load when using it in balanced bridged mode? No. But you will quickly run out of steam, i.e. have much higher distortion at a given output power, as the output stage is already assuming a 4 ohm load when you hook up an 8 ohm speaker in the 1st place. So the general advice is, if you value sonics at reasonable spl's (75-85dB), don't do it
What do you do when you have a speaker with a 4 ohm nominal load? You use the ACA as a stereo amplifier and/or you build an amp that can handle that load, like the Aleph J which I feel is the next step up in this logical exercise of amp building.
Best,
Anand.
Last edited:
Anand,
Thanks for the explanation. I am getting mixed information it seems. I run purely bridged mode xlr and some mentioned its ok to use 4 Ohm rated speakers. I did not build with the option to run Stereo.
I am using Martin Logan Motion 40 mainly because they are the most efficient speakers I have. Specs below. What I'm, confused about are the specs, it shows as 4 Ohms but compatible with 4, 6, or 8 Ohm rated amplifiers. ?? What are your thoughts? Here is an older post with my speaker choices.
Sensitivity 92 dB/2.83 volts/meter
Impedance 4 Ohms Compatible with 4, 6, or 8 Ohm rated amplifiers.
Last edited:
Many has reported about the sound when the amp is turned on (charging of the output capacitor). I don't have it but I use a special PSU where the current will build up very slowly.
Did you adjust the bias as described?
I also don't have the strange sound after switch on, but I've installed a cap multiplier inside the ACA cabinet which ramps up the power slowly. The idea behind the cap multiplier is channel separation - 100dB I believe. And if you were to ask me, I'll say that it was worth it. The soundstage and imaging is exceptional.
.
Ok....and the heatsinks gets hot? …..about 45-50 C?
Next step would be to look at a test signal (after ensuring DC power looks good). It requires a scope to look at and a test generator and an e.g. 8 ohm power resistor as load.
I use a LCLC PSU for each channel (picture). The chokes ensures a slow built up. I have less than 1 mV ripple at 1.4 A. It is 100% silent and the sound is very good. Each choke is a double choke so I can couple them symmetrical for better noise cancellation.
