previous post

TL; DR: There was the loud static noise coming from my Yaesu FT-2900R rig at full volume when I put it on standby, lasting until I either turn it back on or fully cut the power. To fix it, I replaced a JFET that drives the power MOSFET.

Seems there’s some interest gathered around this, so I figured I’d give an update.

I’d like to give credit to Yaesu’s impeccable documentation. Its circuit diagrams as well as the component maps and list provided me with all the information I needed to gain a complete understanding of this device. I did had to look what some of the components do, but it’s all available out there.

I’ve came to appreciate the simplicity of the design. Save for the microcontroller and an EEPROM, it’s all analog.

Anyway, without further ado…

Analysis

I seem to have misunderstood some things about the power management. Seems that during the standby mode, everything but the microcontroller gets turned-off. I was right that the microcontroller would enter the low-power mode.

The 13.8v DC main power goes thru only three different paths:

• the final amp, as its bias, probably connected directly due to the sheer amount of current it needs in its full capacity
• the microcontroller and its supporting components, regulated into 5v
• the power MOSFET, which acts as the main gatekeeper to the rest of the system (audio, VCO, signal processing)

Hence, what signal goes where is completely irrelevant in this state, since they’re all supposed to be off, including the shift register I suspected to be the cause of trouble.

One piece that was key was that the power MOSFET is driven by a JFET (DTC144EE, a p-channel, marked Q1014), which is driven by the microcontroller in a signal line called REG SW (probably means “regulator switch”). The MOSFET’s gate is pulled up to the 13.8v by a 100k ohm.

Observation

Contrary to the expected condition, the power MOSFET kept on delivering power even in standby mode. Furthermore, reading the voltage of its gate yielded 5.9v, which contradicted the supposed 0v in the circuit diagram. What’s even weirder was that the voltage at the gate became higher during standby, at 6.4v.

Since the signal comes from the microcontroller itself, I checked the voltage thru-out the path, where I found that the voltage in the entire signal line was the same as the gate, even up to right at the microcontroller pin. This is especially strange, since the signal level is supposed to be 5v at most.

For good measure, I checked that the resistor values are correct.

Last but not least, I checked for continuity between the MOSFET gate and the REG SW line, and they’re indeed connected directly.

Interpretation

My hypothesis was that the weird value on the REG SW signal line was a combination between the current coming from the microcontroller pin and the main DC power.

With the fact that the MOSFET gate and the REG SW line was connected, and that the resistor values were correct, pointed me to the driver JFET.

Solution

I cannibalize a spare DTC144EE JFET from a scrap. After unsoldering the old problematic one from my rig, I soldered in the good one.

After cleaning up, I connected the power to my rig and put it on standby. To my surprise, the power MOSFET output 0v. For good measure, I made sure the rig was able to do all its normal function (it did).

I then put the rig back on standby, and left it overnight. So far there’s no static noise coming out of it at all.

I guess my fix really works! Perhaps the shady seller put in a wrong component or something.

Continuity

This is definitely not the end of the journey. In fact, it may have only just begun.

One question remain on how the clock signal gets generated, since the microcontroller doesn’t seem to be the one doing the generating after all. I don’t really feel the urgency to really loom for an answer, tho. I’m just glad I was able to get my rig to work!

Anywho, thanks for reading so far. There was actually a whole list of fixes I did to this thing if you care to know more about it.