diyAudio (http://www.diyaudio.com/forums/index.php)
-   Chip Amps (http://www.diyaudio.com/forums/chip-amps/)
-   -   Anyone Care to Explain the Gain-setting Resistors on the Lm4780 Datasheet, Bridged (http://www.diyaudio.com/forums/chip-amps/216757-care-explain-gain-setting-resistors-lm4780-datasheet-bridged.html)

 ZLyzen 25th July 2012 11:26 AM

Anyone Care to Explain the Gain-setting Resistors on the Lm4780 Datasheet, Bridged

So I'm looking over the datasheet for the lm4780 and looking at the bridged example they provide and wondering why the gain was set the way it was.

By the looks of it: the top is the non-inverting part, with a gain set to ~9.87 by the 46.4kΩ and 4.7kΩ resistors and the bottom part is the inverting section with a gain of ~10.87 set by the 51.1kΩ and 4.7kΩ resistors.

I thought each side was supposed to have the same gain, what am I missing here?

Also, is there a better way to achieve the "right" gain for each half without such strange resistor values?

Thanks!

 Osvaldo de Banfield 25th July 2012 12:24 PM

NO!!! the circuit is right. What happens is that one amp is inverting, the gain is -rf/ri being rf the feedback resistor, and ri the input one. The other is noninverting, and its gain is rf/ri +1, so the gain is different for the two arms.

 Osvaldo de Banfield 25th July 2012 01:12 PM

Re-looking at the data sheet you linked, I suggest to make in both channels, the parasitic oscillation stopper (the inductor in series with speaker) particularly if the speaker uses long cables to the amp.), as the fig 5 suggest.

 ZLyzen 26th July 2012 03:19 AM

Okay thanks! It's been a while since I've calculated it mathematically and you're right, they both do have the same gain.
This is actually for a plate amp for a sub so the wires will be exceptionally short, i'm hoping this will help keep things less complicated.
Additionally the load will be 8Ω which isn't too low for parallel, correct?

Thank you again! and so the idea for parallel is to have each channel's gain as close to the other's value as possible correct?

 djoffe 26th July 2012 03:23 AM

National/TI makes a pretty misleading data sheet. There are certain configurations that are simplified, to show principles, but that will get you in trouble.

Figure 2 is one of them. The data sheet says the amp is only stable for gains higher than 10. Well, the bottom amp in figure 2 is set up for a (noise) gain of around 2 if the input is disconnected. That says that the bottom amp will quite likely oscillate if the preamp is disconnected.

There are lots of niceties (input RF filters, output capacitive isolators) that are left off some of the examples. I recommend that you post your schematics to get comments so you can avoid funny problems from conceptual circuits where stuff is missing.

Update My Dynaco

 ZLyzen 26th July 2012 04:01 AM

Hmm, would I be able to get around this by using op amps to convert my signal to a balanced one and basically treating it as a stereo amplifier and just changing the output area?

 Osvaldo de Banfield 26th July 2012 12:22 PM

Quote:
 Originally Posted by ZLyzen (http://www.diyaudio.com/forums/chip-amps/216757-care-explain-gain-setting-resistors-lm4780-datasheet-bridged-post3104838.html#post3104838) Hmm, would I be able to get around this by using op amps to convert my signal to a balanced one and basically treating it as a stereo amplifier and just changing the output area?
Yes, it is possible, but not necessary.

 djoffe 26th July 2012 07:41 PM

This link, and the document referred to on the page may be of some help...

Bridge Mode Beast

 ZLyzen 27th July 2012 01:43 AM

I see what you're saying, so would I be better off running the lm4780 in parallel instead of bridged mode, even though it is an 8Ω speaker? do you personally think I'd get better results?

 djoffe 27th July 2012 03:15 AM

Well, I buzzed the data sheet pretty quickly...It looks like you could squeeze about 100 Watts in bridge mode into an 8 Ohm load, but I'm sure that assumes a very well regulated power supply and excellent heat-sinking. That's more than you could get into 8 Ohms in a paralleled situation, which would be about 60 Watts max. However, if your speaker were a bit less than 8 Ohms, the bridge mode might start to self-protect...it depends upon the frequency, the power supply, and the heatsinking...

Why not go wild? Use 2 of the beasts...a total of 4 amps...run in bridge, using two paralleled amps on each side?

All times are GMT. The time now is 03:35 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.