cool stuff. If you look at the closely transistors themselves (the gold sections) the Base is connected on the left, the Emitter is connected vertical gold bar on the right (connecting to the horizontal gold bars on the top and bottom), and the Collector is on the bottom of the die and its tab (extreme right) actually crosses under the Emitter's vertical bar. Both the the driver and the final look like multiple transistors paralleled and will have ballasting resistors (on die) in series with the Emitters to balance the load across the paralleled transistors. I really like the "slide to tune" capacitors. Most likely so they could use the same module for different frequencies.
Yes, same module, with an easy to adjust tuning, so you make the base module for a wide band, and only have to change tuning capacitor value, and precise location, per batch to get all to be a good match in the desired band. More because the thickness of the ceramic substrate will vary, along with the exact dimensions, per batch you mix, as the shrinkage in firing can be pretty high. Helps then to be able to have a test jig you slide the capacitors along per stage, so you can set the location too place them in production for this particular batch. Look at another, from a different batch, and the capacitors will be in a slightly different location, so if you replace a cracked capacitor in one use the exact same size capacitor, in exactly the same location, or else the tuning will be off.
When describing the RF path on the MODULE, you said (on both transistors) that the RF enters at the EMITTER on the transistors. That is not correct. The emitters are GROUNDED. The RF is applied at the BASE.
You said input to the emitter but then corrected yourself. I see 9 bond wires on the output, is there 9 transistors in parallel then ? Would some of the caps be silver mica? RF is another world , especially GHz, that is RF plumbing with E fields and H fields weird stuff, but I love it. Keep up the great video IMSAI man.
No, all those capacitors are ceramic, a mix of NPO types for the tuning, and for coupling much cheaper and larger value X7R or equivalent for this, as you do not need to worry much about the actual value, providing it is much larger than the tuning ones. Multiple bond wires to keep the inductance down, as you are effectively using the large number of thin wires to create a flat ribbon, which at RF has a much lower inductance. Yes the RF transistors are multiple small structures on the same die, much more than 9, likely at least 18, as each bond wire will feed at least 2 tiny little sections, if not more, with some wide metallised silicon providing a common pad for the base and collector junction, isolated by either oxide, or a reverse biased internal diode, or by having a very high resistance silicon area under them, or a combination of all 3, to keep the capacitance of the connection to the emitter, the base of the transistor, soldered to the heat spreader, down to a very low level. RF transistors are in general more a very specialised integrated circuit component, rather than just 2 junctions, especially when you want high power, small size, and high efficiency at high frequency, all in the same package. But bonding multiple thin wires is a lot easier than trying to bond a ribbon, and have it equally good all the way across the width, plus much more expensive to make a machine to do this, plus the force will very likely crack the die. There are some applications where this is done, but those units are in general a lot more expensive to produce, and tend to use a solder preform both sides to attach the ribbon, when doing the entire die. Not going to be done on a $10 module for sure, probably only on ones that are $1k plus.
cool stuff. If you look at the closely transistors themselves (the gold sections) the Base is connected on the left, the Emitter is connected vertical gold bar on the right (connecting to the horizontal gold bars on the top and bottom), and the Collector is on the bottom of the die and its tab (extreme right) actually crosses under the Emitter's vertical bar. Both the the driver and the final look like multiple transistors paralleled and will have ballasting resistors (on die) in series with the Emitters to balance the load across the paralleled transistors. I really like the "slide to tune" capacitors. Most likely so they could use the same module for different frequencies.
Yes, same module, with an easy to adjust tuning, so you make the base module for a wide band, and only have to change tuning capacitor value, and precise location, per batch to get all to be a good match in the desired band. More because the thickness of the ceramic substrate will vary, along with the exact dimensions, per batch you mix, as the shrinkage in firing can be pretty high. Helps then to be able to have a test jig you slide the capacitors along per stage, so you can set the location too place them in production for this particular batch. Look at another, from a different batch, and the capacitors will be in a slightly different location, so if you replace a cracked capacitor in one use the exact same size capacitor, in exactly the same location, or else the tuning will be off.
I've fixed the 2m band power module on my old Alinco DR605 after I watched your video. Thank you !!!
loved the way the tune the amplifier in production by sliding the caps!
That was cool to see! I ordered this output module for my Alinco - should be here in a couple of days.
very good. i like your use of GIMP
Bring out the GIMP!
ruclips.net/video/S8kPqAV_74M/видео.html
Nice infos..and nice new logo at end of video...Dinos 73
A diode in series with the RF output? Why has that been put in?
TX/RX switching
When describing the RF path on the MODULE, you said (on both transistors) that the RF enters at the EMITTER on the transistors. That is not correct. The emitters are GROUNDED. The RF is applied at the BASE.
Morning again
You said input to the emitter but then corrected yourself. I see 9 bond wires on the output, is there 9 transistors in parallel then ? Would some of the caps be silver mica? RF is another world , especially GHz, that is RF plumbing with E fields and H fields weird stuff, but I love it. Keep up the great video IMSAI man.
No, all those capacitors are ceramic, a mix of NPO types for the tuning, and for coupling much cheaper and larger value X7R or equivalent for this, as you do not need to worry much about the actual value, providing it is much larger than the tuning ones. Multiple bond wires to keep the inductance down, as you are effectively using the large number of thin wires to create a flat ribbon, which at RF has a much lower inductance. Yes the RF transistors are multiple small structures on the same die, much more than 9, likely at least 18, as each bond wire will feed at least 2 tiny little sections, if not more, with some wide metallised silicon providing a common pad for the base and collector junction, isolated by either oxide, or a reverse biased internal diode, or by having a very high resistance silicon area under them, or a combination of all 3, to keep the capacitance of the connection to the emitter, the base of the transistor, soldered to the heat spreader, down to a very low level. RF transistors are in general more a very specialised integrated circuit component, rather than just 2 junctions, especially when you want high power, small size, and high efficiency at high frequency, all in the same package.
But bonding multiple thin wires is a lot easier than trying to bond a ribbon, and have it equally good all the way across the width, plus much more expensive to make a machine to do this, plus the force will very likely crack the die. There are some applications where this is done, but those units are in general a lot more expensive to produce, and tend to use a solder preform both sides to attach the ribbon, when doing the entire die. Not going to be done on a $10 module for sure, probably only on ones that are $1k plus.
@@SeanBZA Wow. That was very educational. Thank You Sean.