Darn mate, you are really selling the 2x ACA idea😀 Im kinda reluctant, as going the flea power SE approach was something I "idolise", but than I have no experience with the balanced approach and I fully belive your (and others' claims). The second thing, im in Europe, and this second ACA will be another 600$ after shipping, customs/tax. Than I'm thinking, I might as well try to build a F2J.
But I have tried almost every position in our 30sqm-ish living room, even the less practical ones and I have landed on one (speakers and listening position) where modes are working rather favourably. With ACA now, Im only missing maybe 3-4db low-mid bass (on good recordings). I know this when i put +3db low shelf from 125hz on Mojo2 dsp. Its only the upper bass/lower midrange (150-500hz) thats a bit lacking compared to the hot upper midrange (500-1500hz), that makes the overall sound a bit lean/dry.
I get similar effect when i use 5m (15ft) of a single twisted pair of cat 5 cable (so 10m total of 26awg solidcore). I can measure the resistance of it even with a multimeter at ca. 2.1 ohm i guess a good approx. But that wire just spooks me with two small kids jumping around all the time, so im back to 5m Belden 16awg (the red/black twisted and tinned one, which is probably closer to 8m untwisted, popular among vintage crowd) + 3db low shelf on the dsp.
I guess trying a conservative passive network as discussed with allen and GM will be the thing to try first, than do some real measurements.
Im assuming a paralel (zobel) network will do very little with only 1.5ohm Rout of the ACA?
I can tell you that Mojo2 drives the fostex horns very loud without distorting at all, on most recordings. Its only some classical recordings that prove a bit too much for it. The fostex are rated at 96db though, and Chord head-amps can apparently deliver quite alot of current, even on a battery. Mojos engineer Rob Watts even said he designed it with driving speakers in mind and he is using a combo of m-scaler with TT2 to drive his personal BW speakers, which are not efficient at all! Its rather impressive, i must say.
Pair of ACA is a budget, reliable flea power amp with a bit more than flea watt power (10dB more than a 45 or EL84 [trioded]).
dave
I would definitely recommend trying an F1 or F2 of any flavor before committing to massive driver or cabinet mods, just so you know what you can get that way first.
I've got a pair of Fostex FE126Ens in the BK-12m folded horn enclosures. I've played them with my ACA, both in stereo and in bridged / pseudo-balanced* mono, and with various other amps. The ACA seems a little narrow and undefined to my ears relative to my F3, my F1J clone, and Zen Mod's Redneck DEFiSIT when driven properly. The FE126Ens really come alive with all three of the latter amps driving them.
*this is a preamp problem, not an ACA problem, rather not get into it, it's complicated, &c. 😉
Last edited by a moderator:
Both of these amps take you further into current amp territory, F2 has Rout a bit over 2x bridged ACA, so moves from the lowest of the suitable Rout to the highest. F1 is abourt 10x the F1, decidedly on the current side of the line, you might all of a sudden find you have excess bass (not low bass, you horn is incapable of that, but strong midbass.
Both designed to be higher quality than ACA, a major driver of its design, the ability for a punter to build one in an afternoon at an Amp Camp.
dave
Yeah. The factory F3 Zout is 1.3 ohms, very close to an ACA in balanced or bridged mono (1.6 ohm), so it might be a good apples-to-apples comparator to the ACA.
Last edited by a moderator:
I've wondered how much of that is power supply. Played soft the ACA has a lot of detail and delicacy. I've been toying with the idea of powering it with the F1J's power supply, each channel would then have the benefit of it's own 200VA toroid. But the F3 would still have the advantage of a rail at twice the voltage.
Last edited by a moderator:
For the beginning, try a simple trick: connect a hi-wattage resistor (25w or so) 20-40 ohms in series with the + lead. Yes, it will kill the efficiency, but listen a bit like that and then you will have a better idea if you will need a "current amp" (F1 or F2) or not...
Yeah, I use to recommend putting a cheap 25 ohm pot in series with each speaker to 'dial in' the desired response and sub a fixed resistor at whatever the pot setting or just buy quality pots, budget permitting, to mimic a high power tube amp of days gone by for all but the luckiest or wealthiest among us.
Thanks Vix & GM, thats exactly what I was thinking, a pot might be the most informative as one can quickly test various values. One question regarding this, can one expect the same result (frequency response) with an amp that has Rout of 8ohm (ie. F2J) and am amp with Rout 1.5ohm + 6.5 ohm resistor in series (ie. single ACA+resistor) - neglecting power handling ofc?
In terms of a voltage divider, wouldnt they be the same? Why use such high values (20-40ohm) than?
In terms of a voltage divider, wouldnt they be the same? Why use such high values (20-40ohm) than?
Tweaking the ACA is on my list, somewhere down the line. I've browsed through that immense "upgraded ACA" thread, but that'll take a year to get through. I was thinking if an off the shelf external LPSU could work just as well? Seems the easiest, just plug it in and also keep it far away from the circuit?
Not sure what sort/spec would be suitable. Likely needs to be quite a bit more than SMPS 24v right? Or is this idea rubbish?
I wouldn't change the voltage spec, personally, not at first anyway. The rationale or why for a different PSU, specifically a linear one, is that it should have more current reserve and less rail sag under demanding program material, and hence may sound more dynamic or full. This is speculation, I have yet to test or confirm it. While I also believe that a higher rail voltage and/or Iq does affect the perceived loudness and has the potential to hit a different sweet spot of lower distortion with a different operating point, I wouldn't want to jump directly there. Too many changing variables means identifying root cause for any perceived improvement can't be done in one experiment.
Regarding the how. I suggested repurposing an existing F1 style PS out of personal convenience and because it would be 'known good' having come from a working amp. If you've never built a mains supplied PSU before I wouldn't recommend this as a first try, due to the risk of error and electric shock and inability to confirm your work independently. I'd want to build an entire amp that needs one first, like one of First Watt or Aleph clones that us folks here like to fool with, and get that working, then borrow the supply. But that's asking a lot. Maybe make a fren who has one, & borrow it.
I don’t know if we can say that. Need to talk to an amp guy and ask how the output impedance changes with frequency & dynamics. The resistor is pretty close to fixed.
But yes, approximately.
dave
Last edited:
You're welcome!
What Dave said. 😉
Not an amp guy, but thanks to Altec I can show how variable DF tone controls impact frequency response and knowing I can't imagine boosting beyond ~ +10 dB plus factoring in that at least at RadioShack, 25 ohm pots were cheap due to so many being sold it's what I've used/recommended so not sure ATM why up to 40 W would be needed.
In short, to my ears when wanting to get the best performance overall out of a FR driver, then the pioneer's tube receivers is the way to go and for multi-ways, a Class A amp for the woofer and SET tubes for the rest since tube LF sounds a bit too 'syrupy' (lush?) for lack of a better description.
Primarily yes.
Although a possible non-linear output impedance has been mentioned, it is only of secondary importance with regard to the task of equalising response via the impedance. It is also unlikely to be of much significance in this regard.