Any advantage to make dual mono (PSU) for the gainclone (I'm ordering Brian GT's kit)

[Peter Daniel]

Originally posted by metha
IMHO, running the chip at 50W may be too high, the SPIKE protection might be activated, partcularly, if the speaker impedence drop under 8Ohm at some frequncy (re: the datasheet).




uhm.... , the price of speakers is expensive, that's why I prefer to be on the safe-side. Do you mean if there is a coupling cap (1.0-4.7UF) it's safe/very low possibility to have dangerous DC voltage at output (even we also obmit the cap in the feedback loop which ensure DC untiy gain) ??!

We run Amp-1 at 34 V on rails (and max allowable is 40V), so it is not really stressed. I find it as a good compromise between 25W and 50W, how much it really is I don't really know. The point is it plays well and is well balanced. Specs were never important to me. If it's way out of line, blame marketing dept. not me

Any offset below 100mV should be pretty much safe. The max I've seen on those amps was 80mV and I'm not loosing sleep over that. If it breaks your speakers, we might buy you new ones (but again it's up to marketing dept.). The input coupling cap protects the output if source components pass DC to the amp. The good ones shouldn't. I don't believe those amps suddenly break and damage the speakers. They have more protection circuitry than some other well regarded class A amps ( which also don't use protection circuitry).

I had the output of my amp shorted for at least 1/2 hr and it didn't cause any damage, the amp just got rather hot, but nothing else. There is also built in protection in case one rail goes off.

[metha]

Peter - Thanks for the clarification. :-)

[Nuuk]

There is also built in protection in case one rail goes off.

Are you sure Peter? I had a wire break that supplied the negative rail and I got the full negative rail voltage across the speaker terminals!

[Peter Daniel]

I once had a bridge failure and this didn't produce any DC at the output. I also recall analog_sa mentioning testing that possibility with some crappy drivers and it seemed like DC offset didn't occur at the output (when one rail was off). But I didn't really tested it myself, will try it maybe today. I also think that you need load to be connected to observe it, otherwise there will be DC.

Here what's mentioned in Datasheet about protection circuitry:

GENERAL FEATURES
Under-Voltage Protection: Upon system power-up the
under-voltage Protection Circuitry allows the power supplies
and their corresponding caps to come up close to their full
values before turning on the LM3875 such that no DC output
spikes occur. Upon turn-off, the output of the LM3875 is
brought to ground before the power supplies such that no
transients occur at power-down.
Over-Voltage Protection: The LM3875 contains overvoltage
protection circuitry that limits the output current to approximately
4Apeak while also providing voltage clamping,
though not through internal clamping diodes. The clamping
effect is quite the same, however, the output transistors are
designed to work alternately by sinking large current spikes.
SPiKe Protection: The LM3875 is protected from instantaneous
peak-temperature stressing by the power transistor
array. The Safe Operating Area graph in the Typical Performance
Characteristics section shows the area of device
operation where the SPiKe Protection Circuitry is not enabled.
The waveform to the right of the SOA graph exemplifies
how the dynamic protection will cause waveform distortion
when enabled.
Thermal Protection: The LM3875 has a sophisticated thermal
protection scheme to prevent long-term thermal stress
to the device. When the temperature on the die reaches
165°C, the LM3875 shuts down. It starts operating again
when the die temperature drops to about 155°C, but if the
temperature again begins to rise, shutdown will occur again...

[net-david]

Originally posted by Peter Daniel


We run Amp-1 at 34 V on rails (and max allowable is 40V), so it is not really stressed. I find it as a good compromise between 25W and 50W, how much it really is I don't really know. The point is it plays well and is well balanced. Specs were never important to me. If it's way out of line, blame marketing dept. not me

Looking at the data sheet, it seems that the SPiKE protection will kick in at about 55w into 4 ohms with 27~28v rails. Since I know my speakers have a minimum impedance of 4 ohms, wouldn't I be better off shooting for the lower rail voltage? It looks like I would still acheive 35w into 8 ohms.

David

[dw1narso]

Halo Nuuk,

terima kasih atas reply pertama kali dari anda.

Are you Indonesian tinggal di UK? or you are an Englishman speak Bahasa Indonesia?

Actually I speak Jawa also (he, he, he it is really Javanese language not JAVA, that coffee cup programming language).

Brian,

about the 1KVa transformer, isn't it to highly regulated? for just a 50 Wattage amp? I read that too regulated PS will slow the amp...?

And another question is...,

if I cannot find any toroidal transformer (must import to get this kind of transformer in Indonesia), will EI transformer be sufficient?

Thanks,
---
David

[grataku]

Originally posted by Peter Daniel

It is Zebrawood and it gets much nicer when sealer (or oil) is applied. This Patek indeed deserves it's name, it is so cute and features almost watchmaking precision

But it's possible only with those chip amps. Rodd was listening and didn't dismiss it

I always confuse zebrawood with tiger maple! I certainly would entertain the idea of auditioning the amp in my system. The summer is coming and the AX will get retired for the season.

Metha,
what's your problem with my username?