LM4780 single supply circuit

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
W

WielkiElektron

Member
Joined 2015
I am building single supply amplifier using LM4780. I have found Figure 5 in the datasheet which presents a single supply amplification circuit. I would like to make sure if I denoted vee, gnd and mute properly. I would greatly appreciate any comments.
 

Attachments

  • makesure.png
    makesure.png
    42.4 KB · Views: 291
sgrossklass

sgrossklass

Member
Joined 2004
stephan.win31.de
Looks correct to me.

Note that Ci is by no means "optional" here, and R_SN and C_SN are very much recommended as well. Q1 also is a generic type and better supply rejection may be obtained using a part with higher Early voltage at moderate Cob (BC556B maybe, or 2SC2240, or a small TV vertical deflection transistor if you happen to have such floating around; it shouldn't be too big since current is only ~2 mA max).
 
sgrossklass

sgrossklass

Member
Joined 2004
stephan.win31.de
Speaking of output caps, I'd consider establishing a second (AC-coupled) feedback path to the output side of the 4700µ. Maybe 24-26k and 1-3.3 µF, while increasing Rf to 100k, or something along those lines. Historically some designs have also split resistance 50-50, which means you're only getting rid of half of capacitor distortion. It's probably a tradeoff between minimum distortion and stability (keep in mind the output cap has some parasitic inductance, though one could counter this by paralleling something like 220µ and 2.2µ).

A split supply design tends to take up less room, of course. You may be able to get away with half the capacitance at twixe the rated voltage for supply buffering, but you don't get around big output coupling caps.
 
Last edited:
blu_glo

blu_glo

Member
Joined 2005
WielkiElektron - Yes I would concurr you have that correct.

It sure does appear from the diagram that single rail support was almost the last thing on the IC designers mind... all they have done in above is just set up a low impedance psuedo- 0V for the op amp to continue working correctly from single supply. There have been other chips that could do this automatically needing only input bias (could be medium impedance) and output coupling cap.

Just out of interest, why do you want to use the single supply option?
You save nothing as you still need 2 large electrolytics and now one is in your signal path possibly adding distortion. (sgrossklass's suggestion would indeed remedy that, to a degree).....
 
Last edited:
W

WielkiElektron

Member
Joined 2015
blu_glo said:
WielkiElektron - Yes I would concurr you have that correct.

Just out of interest, why do you want to use the single supply option?
QUOTE]

Since 100W 48V switched-mode power supplies are cheaper than toroid transformers. I’m going to apply one of desktop power supplies which looks very similar to typical notebook power adapters. It costs less than 15$. Moreover, I can use smaller beautiful aluminum enclosure and I do not need to bother with a protection against an electric shock which may cause 230V. I like sound of LM4780. I have already used it in the typical application circuit depicted in the datasheet. Although this chip is less efficient than D-class amps, I think that LM4780 is an interesting choice for not laud listening to music. If I verify that my prototype with single supply application sounds pretty as good as the typical application with dual supply rails then, perhaps, I will considering launching my own small business and producing amplifiers with the use: LM4780, PGA2311 for a level control, IR receiver, ATMEGA for a control, and 3 relays with ULN2803 for an input selection. I recon there is a small gap in the market for very cheap (60$ or so) and good sounding amplifiers with an input selector and a remote control. I would prefer to use a dedicated chip for the precision virtual ground similar to TLE2426 instead of a transistor but, unfortunately, I am not able to find a voltage splitter chip with an input voltage greater than 40V. I am sorry for my poor English.
Click to expand...
 
A

AndrewT

R.I.P.
Joined 2004
You will need to investigate what peak or transient current the smps can output without shutting down.
You will need to investigate what output (smoothing) capacitance that the smps can charge up at start up without shutting down.
You will need to investigate what effect the high frequency supply line ripple will have on the chipamp performance.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom