Low-distortion Audio-range Oscillator

[dirkwright]

Originally Posted by rsavas

Dirk,

Putting bypass caps on the opposite side of where the part lies is not recommended for good high freq decoupling. Vias create problems as they have inductance, something that you do not want for good de-coupling. Maybe the larger bulk decoupler (1uF) may go on the opposite side, but the smaller ones should be on the same side as the IC. The layout rules usually goes to put the smaller cap closest to the pin followed by the larger ones. Say 1nF,100nF,1uF. Using TH caps is not recommended as they take up a lot more space than a SM caps do, leading to more trace inductance. Choose X7R dielectric for the big ones and COG for the smaller ones, is usually recommended.The trace is to flow through the caps hi side, then finally to the IC pad/lead, with minimal trace length, in other words, no stubs. 0805/0603 are great for this. I agree, not that easy for a bread board design or discrete wiring prototype.

Rick

OK, thanks. I did not know that. This makes it even harder to use this SMD opamp because it's so small that it's hard to get all the SMD resistors and other components really close to the pins if there are power supply bypass caps in the way. I don't think the LME49990 is good for DIY design, even if I design my own boards. I'm going to try though. This is getting off topic though.

[rsavas]

Lets not generalize if a part is good for DIY or not, a lot depends on the design of the PCB, regardless what the application is. Yes you can argue that some DIY's are not well versed in layout or SMT soldering, but that can also apply to many seasoned EE's in a professional capacity. It really depends on your experience, regardless of your credentials.
Rick

[dchisholm]

Originally Posted by rsavas

The layout rules usually goes to put the smaller cap closest to the pin followed by the larger ones. Say 1nF,100nF,1uF. Using TH caps is not recommended as they take up a lot more space than a SM caps do, leading to more trace inductance. Choose X7R dielectric for the big ones and COG for the smaller ones, is usually recommended.The trace is to flow through the caps hi side, then finally to the IC pad/lead, with minimal trace length, in other words, no stubs. . . .

Rick,

By any chance can you suggest some appropriate mfgr's part numbers, and a suggested PWB layout sketch for power supply decoupling in an application like this? I vaguely recall some other forums where somebody tried to use the LME4990 and had stability problems. They concluded it was interaction between the opamp's inherently high gain-bandwidth product, and the PWB layout. Would be helpful to see a layout geometry that's known to work.

Dale

[KSTR]

Originally Posted by KSTR

looks like Samuel was right and the shown results were unreal in terms of absolute levels. The synced averaging and the AP graphing methods are to be taken with a grain of salt.

I now finally figured how this works with only little attenuation of the spectral lines, although on the AP2322 (Sys2) this time -- won't matter much. Damn Win7 won't run the old APWin for some reason, so I post this next time I'm on XP again. I also measured the close-in behaviour of Viktors' oscilllator, no problem there that I could identify, will post that, too.

[rsavas]

Hi Dale,
Look at Owen's thread,
http://www.diyaudio.com/forums/headphone-systems/179298-wire-ultra-high-performance-headphone-amplifier-pcbs.html
For parts, there are litterly 100's of possibilities, go to Mouser and search/filter for
capacitor-smt-mlc-50V-1u-X7r
capacitor-smt-mlc-50V-100n-X7R
capacitor-smt-mlc-50V-10n-NPO(COG)
Owen's threads have part #'s (BOM's)
A number of IC mfg have app notes on hi-speed layout design techinques. I know TI /Intel have many. I will attach some that I have sitting on my jump drive.
You can also look at my thread for my layout of the LME49830 as well,
http://www.diyaudio.com/forums/solid-state/220962-lme49830-latfet-yet-another-design-review.html
and Owen's LME49830 LatFET design,
http://www.diyaudio.com/forums/solid-state/196932-wire-amp-class-ab-power-amplifier-based-lme49830-lateral-mosfets.html

[rsavas]

Some more

[rsavas]

I have attached pic of the LME49830 layout, showing the PS decoupling, hope this helps.
Ground planes are so important, in some of the hi-speed design layouts we did in the past, we used NELCO material and stacked up the PCB, having very small distances between planes, so that they acted as distributed capacitance layers, sandwich power planes between ground planes. One design we used 16 layers with a number of embedded power/ground planes. This was done in 0.062" total thickness for a PCI FF PCB. I'd have to look at the stackup again to find the material specified, but you should get the idea.
Rick

[RNMarsh]

Originally Posted by RNMarsh

thd is high -- dc specs are very good. Neds to be included within gnfb loop.

never mind - thinking of different P/N.

[Samuel Groner]

Related to this topic (no worth starting another thread): I've had the chance to scan the manual of the calibration fixture (067-0938-00) which was used to verify distortion performance of the Tektronix SG505. The PDF is made available here:

067-0938-00 - TekWiki

Good inspiration for those interested in building a passive notch filter!

Samuel

[1audio]

That's a huge file for a manual. Thanks for making it available. It illustrates the complexity necessary to make something that is flexible. The switching looks like its similar to a Tek attenuator switch.