Elliott Sound - "The Project 3A"

Status
Not open for further replies.
Jump to Latest
#3
Hi!

I just finished the latest evo of my P3A yesterday. As always, no digicam available, so still no pics...

I used MJ15003/15004 and BD139/140 as transistors.

I exchanged C2 (220pF) ceramics for Mica (Mica way better than anything else), C7, C+ and C- (100 nF MKT) for MKP-type, C1 for 4.7 uF MKP and C4, C6 (100pF ceramic) for FKP

No other changes. I used a semi mono supply design (one big toroid - 500VA 30-0-30 - and two seperate supply boards), and also split the original P3A board in half. For rectifying I used four MUR860 disodes, and for supply caps I used 2 * 10.000 uF BHC SlitFoil for each channel, so 40.000 uF and 8 diodes at all (and R. Elliot himself has underlined it time and again that more capacitance really would be a waste of money... he even told me to use a 300VA transformer instead of the 500VA one, but since that only costs 8€ more than the 300VA type...)

I heatsinked (very small cute TO220 sinks) all the BD transistors, so that in spite of Elliots recommendations I should be able to run the amp into 4 ohms with the 42 V supply voltage I use...

Well, how does it sound? I am not an expert at describing sounds, and the only stuff to compare it with 🙂cool: 😎 😎 ) are a GainClone (OPA548) and a LCAudio End Millenium module...

I cannot really spot a lot of differences between the GC and the P3A, but of course P3A seems more powerful when it comes to high volumes (actually my speakers are 6 ohms), and does not clip as fast (but I only can get all amps into clipping by adding a pre-amp, normally I am running all my amps without a pre-amp)

No noise (only some hissing from the tweeter which gets unhearable when more than 10 cm away from the tweeter), and overall a neutral, if slightly warm sound (a lot warmer than the LCAudio End Millenium, which must be the most neutral amplifier ever build, therefore I normally only use it as monoblock for my subwoofer).

Overall I must have spent about 140 € (including everything, also case and a vandalism proof push button and a solid state Finder relais). About 40 - 50 € more than for my GC, but since now I like the "massive" appearance of the P3A (19" case) more than the very small GC, and I also plan to use 83dB speakers in the future, for which the GC would be a little underpowered...

Bye,

Arndt

PS: As for power, I estimate (no way of measuring it) my amp at about 90 - 100W per channel into 8 ohms. Take a look at Anidin's site, there is a power calculator for the P3A...
P3A Construction Guide
 
#5
thamks for all to the answers.....

sayang001..... how u can get 180W RMS with 60+/- and in the ESP site thay say that 90W RMS MAX on 8 ohm speakers.....
u are change something ? if it is yes..... so what u change..?
u can post your circuit ?
 
#8
sajti.......

if i build this amp with my part...... and each amp us in 6 O/P devices...... of 2sc5200/a1943......how much power i can get in 4 ohm ? and 8 ohm ? ( with +/- 68v ) and i need to change something in the circuit to make him work in this high volt....?
 
#9
Hi DJ,

this is not very esay to increase the output power of any amplifier. If You just add some more output devices it can results oscillation, which can kill Your transisitors. If You connect together more power transistors You have to use some small resistors to equal the current flow thru the devices.
If You want to build high power amplifier just go, and search for high power amplifier schematics!

Sajti
 
#13
DJ,

the 2SC/2SA pairs OK for high power, the MJE drivers also OK for high power. The BC639/640 is OK for input stage. But You need some medium power, high voltage transistor, as VAS (Voltage Amplifying Stage). Such as MJE340/350.

With 6 output devices 200-250W is possible with +/-55-60V to 4ohms, or with +/-70V to 8ohms.

Sajti
 
#15
Hi!

One more thing about P3A: R. Elliot tends to be very conservative (that is: on the safe side) with his recommendations concerning rail voltage, transistors and heatsinking.

I would say, that if you use big enough heatsinks for all power transistors, and also for the drivers, higher rail voltage than 42V (the max according to the P3A construction guide) can be used - or lower load impedance.

Right now I use the MJ15003 / 15004 as power transistors (TO3 devices), and one pair of them is directly mounted (no L-bar) on a 0.6 C/W heatsink. Even after playing about 1 hour at 1/3 max volume, the sinks only get handwarm.

If you want to get even better power handling and/or heat dissipation, maybe you could use the biggest Sanken devices... those badass transistors have a real big surface to the heatsink and can handle a lot of power... but not very cheap...

Bye,

Arndt
 
#19
i build this........!
 

Attachments

  • p68-f1.gif
    p68-f1.gif
    10 KB · Views: 1,056
Status
Not open for further replies.
Top Bottom