fst3253 as 0°-180° experiments failed

[Tonytti]

I tested fst3253 TX_RF solution to get the 2 phases to drive fet's but load for pico is too high and after early 5 minutes that seemed ok, pico turned off, dead. It needs a buffer, but this complicates my semplicity goals. All pin connections were correct and the two outputs were ok, but nothing to do. Ohmic resistance on fst3253 pin 14 (input TX_RF) is in the order of few tenth ohms, so I deduced that it needs more current than pico gpio can give. Any suggestion? Better go with TC4428 buffer driver as one chip solution ?
Thanks for your attention.
Antonio

[dawsonjon]

@Tonytti My suggestion was that the pico would drive the s0 input of the 3253, 1A and 2A would be connected to the gates of the BS170s, 1b1 to 5v, 1b2 to 0v, 2b1 to 0v and 2b2 to 5v. The Pico isn't supplying any significant current in this configuration; the 3253 is doing all the heavy lifting. We know the Pico can easily drive the S0 input of the 3253 because it's used in that configuration in the receiver. My guess is that something else failed first, and the Pico was a secondary failure.

On paper the 3253 has a higher current handling capability that a 74HC14, but the HC series devices are known to be quite robust. You could consider swapping the 74HC14 for a 74HC136 with 4 xor gates. Each XOR gate can be configured as an inverter (by connecting one of its inputs high) or a non-inverting buffer (by connecting one of the inputs low). This would offer a single-chip solution, and you could use pairs of gates to double the drive capability. I think this would be sufficient to drive a BS170.

TC4428 has a much higher drive capability but will struggle at higher frequencies. It looks like UCC27517 would be up to the job, but it is probably overkill for a BS170. If you were going down this route, you might want to consider using something like an IRF510; this could probably give you about 10W using a single transistor but wouldn't be very efficient.

[Tonytti]

Thanks Jon, chip was burnt, in fact 39 ohm from S0 and ground was a sign of failure.
Cheers!

[ea2ehc]

Hi, I finally managed to compile the code for the RP2040, with a few tweaks, and .... it works !!!, so now I can experiment with the transmitter section, in my case (CW) using polar mode with the internal NCO. The idea first is to try it with an FST3253 as you suggested, although I'll try the USDX model, since it has a MOSFET acting as an RX/TX relay.

Thank you very much, Mr. Dawson, for your work.
In the schematic, I think you have the MIC pin wrong; it should be pin 34 (ADC2), not pin 33, which is GND.

Suggestion: Perhaps a chat on group.io would be better for exchanging ideas.