I will add tuning mechanism at the output of U2 inserted between R3 and R7 because this is the culprit in making the duty factor to ~51%. R7 effectively add resistance to R3 during charging of C2 but discharges quicker via output transistor without R7 ugh!
Fig. 2. Circuit showing LM319 with duty factor correction using trimpot
Comparison output of U2 and U1. Blue trace is U2, Yellow trace is U1 at 51% on both.
This is the breadboard layout showing the added TC4420 FET driver. This is a preliminary sound testing that I need to check the quality of the sound it produce before I transfer the complete circuit into a PCB. This FET driver is qualified to drive my stock 5 watts speakers. I`m excited!
Note! There is no LC filter yet.
Here is the result output combined when amplified with FET driver. The blue trace is TC4420 and the yellow trace is LM319 at pin 7 of U1. The input buffer of TC4420 gives a sharp dual threshold voltage that improves the T rise and T fall of LM319. This is always needed to improve the switching efficiency of power FETs.
Very nice output improvement indeed.The blue trace is TC4420 and the yellow is LM319. The spikes at the leading edge of the blue trace is just the capacitance of the test probe– nothing to worry.
When all the needed parts were in placed on the breadboard and test fire this thingy, – Lo and behold! the sound is surprisingly good!
What is my final say?
I can therefore say that generating triangular wave out of a simple resistor–capacitor network can give decent sound quality but at very low cost resulting to an effective design approach that can be reckoned with. Well, at least this is only if you dont want to buy precision high speed comparator online or just simply avoiding to spend much money for your next outdoor audio amp project. LM319 is as abundant as 555 timer IC, so you may add it into your arsenal of chips inline with your tl494 class d amp. If you appreciate the sound of TL494 as class D amp then you will like lm319 aslo.
But wait! the circuit drawing in Fig. 1 and 2 is the same circuit lay-out I made on my breadboard. this means you can expect some ringings on it from your scope due to long jumping wires and resistors leads that will radiate a lot. That lay-out you see on picture is the lowest possible noise I could do but trasfering to good PCB lay-out will give more satisfying sound performance.
Here is the sound that I apploaded in youtube using the built-in mic of my crappy smartphone. https://www.youtube.com/watch?v=fDZIXCRCzvI