M0LTE / pi-transceiver

An attempt to build a complete ham radio transceiver around a Raspberry Pi
14 stars 3 forks source link

Selection of approach for 144MHz power amplifier #30

Open G4CDF opened 2 years ago

G4CDF commented 2 years ago

Initial thoughts were to use a Mitsubishi module (eg RA35H1516M). These are obsolete but easily obtainable. See

http://www.g4izh.co.uk/mitsubishi-ra35h1516m-pa-module-conversion-to-4m.html

for more information.

Data sheet is at

https://www.mitsubishielectric.com/semiconductors/content/product/highfrequency/siliconrf/module/ra35h1516m.pdf.

This is suitable for initial breadboards, but I am not yet sure of its performance, although I have used one in a 2m transverter for several years without issues (once I heatsinked it properly)

G4CDF commented 2 years ago

I then discovered some low cost dual LDMOS devices that are designed for 12.5V operation in Mobile radio service. A suitable final would be AFT05MP075GNR1 from NXP. Price from mouser is £16.47 +VAT. Data sheet is https://www.mouser.co.uk/datasheet/2/302/AFT05MP075N-1126063.pdf This can provide 75W for 144 and 432 MHz (different design of PCB). In fact it is used to provide 100W on the IC9700!
The design looks straightforward if NXP will provide data for 144MHz and possible if they do not. This assumes use in parallel rather than push pull.

G4CDF commented 2 years ago

400W is a dream that will not happen for the following reasons: W6PQL has an excellent 1kW linear design I don't fancy 80A at 12V There are no LDMOS fets that will give 400W at 12.5 V and combining would be a nightmare. Unless someone knows otherwise 100W is a reasonable upper limit. Clearly Icom thought the same.

M0LTE commented 2 years ago

50W would be great. Even 25W would be a big success. 100W and especially 400W is what I’d consider exotic for 2m. An external linear might be a nice follow-on project.

On Tue, 18 Jan 2022 at 11:44, Mike @.***> wrote:

400W is a dream that will not happen for the following reasons: W6PQL has an excellent 1kW linear design I don't fancy 80A at 12V There are no LDMOS fets that will give 400W at 12.5 V and combining would be a nightmare. Unless someone knows otherwise 100W is a reasonable upper limit. Clearly Icom thought the same.

— Reply to this email directly, view it on GitHub https://github.com/M0LTE/pi-transceiver/issues/30#issuecomment-1015332413, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJAQNWFDM7XYDVN4LMOLR4LUWVHCHANCNFSM5MGQ3QTQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

G4CDF commented 2 years ago

Go for W6PQL 1kW design... At 600W it is nicely underrun even allowing for coax losses. At this power 28V or 50V will do for supply.

G4CDF commented 2 years ago

Very sensibly we will go for the Mitsubishi module first: thanks Richard M0SKP