PA-100D Alarm

[c813]

My PA-100D has been working fine since I bought it, but I recently found that in normal STBY standby mode, after a period of standby or sending some cw in OPER mode, PA-100D will alarm, and it cannot be eliminated by PWR power supply. After observation, it seems to be a HI Cur fault, because in normal STBY standby mode, the LED shows 0.1A, but when the alarm is on, the displayed current is greater than 20, 30A, and the alarm sounds. What is the reason for this? What methods and steps can be used to check and determine the fault point?

[5B4AIY]

There are several possibilities for this fault to occur. Attached is a copy of the schematics associated with the over current sensing. I am assuming that this is not a real over-current situation, that in the standby state the amplifier is only actually drawing 500mA to 600mA when the alarm occurs.

The MAX4373 high-side current sense chip detects the voltage across the current sensing resistors and outputs an analog voltage of 0.1V/A which is digitised by the microprocessor and displayed as a current. This analogue voltage is stepped down in a potential divider formed by R19 (10K) and R20(3K3). When the voltage at the junction of R19/R20 exceeds 2.4V this trips the over-current latch sending a +5V logic high to the microprocessor which responds by tripping the over-current alarm.

First step is to ensure your current sensing resistors and the connections to the transistors are OK. With a multimeter measure the resistance between the main power connector and L3. It should be essentially a short circuit. Carefully examine the PC board in this area to make sure there are no cracks or other defects.

Next step measure the resistance of the potential divider resistors, particularly R20. If this seems to be OK, at the next occurrence of this spurious alarm, measure the voltage at the junction of R19/R20 if it is high, above 50mV then there is a fault with these resistors. If the voltage is low, then check the voltage on the O/C line, if it is high, +5V, then this would indicate a probable fault with the MAX 4373 chip itself.

As this is an intermittent fault I would infer from this that it is more likely to be a bad connection somewhere around R19/R20, especially if when working normally the ammeter registers sensible currents, and only shows high current with the fault present but in standby.

Adrian, 5B4AIY

[c813]

Thank you 5B4AIY, thank you for your suggestion, I will follow your suggestion to check, thank you.
Another question: During use, using CW or SSB, the amplifier works and displays normally, but when performing Digi data communication (such as FT8), at the beginning of transmission (TX), the band display position on the LCD screen of the amplifier changes from band to "?". What is the possible fault? Is it the same reason as the over-current fault?

73, thanks.

[5B4AIY]

I hope my suggestions help, let me know. As for your frequency sense problem, that's a new one. When I modified the firmware - I am assuming that you are using my modified firmware - I spent a lot of time on the frequency sense logic as the original firmware had a tendency for 'relay chatter' at the syllabic rate in SSB. I have not had any reports since of any F-SENSE problems, but I suppose it's always possible.

You can start by using the F-SENSE mode, which will give you an extra display page that shows the actual input frequency to a resolution of 1kHz. Using CW, and a dummy load and about 100mW to 1W of drive power, key the amplifier in CW and check that the frequency displayed is immediately correct. This will verify that the F-SENSE logic is operating as desired. Then switch to FT8 and key the amplifier. What does the frequency display do? It is possible that the FT8 input could 'confuse' the F-SENSE logic. To check this you would need to use the TEST mode and check the contents of the 'bins' in the F-SENSE TEST. Run the same test, and observe the contents of the 'bins' when keying CW, and then switch to FT8, key the amplifier and see what is in the bins this time. Using a serial test program you could capture the output and send it to me. I'm not promising any changes to the firmware, but if there is something simple that could fix this problem, then I could see what I could do. Bear in mind that I have not changed any firmware now for several years, and I'm not sure that the old development system will even run on my present computer as well as the Ingenia firmware update software.

73, Adrian, 5B4AIY

[c813]

I did not select F-SENSE mode. I used kx2/kx3 in "auto band detect". The cable in the rs232 socket was a cross-wire of tip and ring. It was normal when used before (including ft8). After the O/C over-current alarm appeared, the ft8 operation began to have problems. The firmware of the amplifier is v4.1a. In the past few days, I also found that when only kx3 and pa100-d are connected, the band always shows "?". When switching bands on kx3, it also shows "?". Only when I open the log software (RumlogNG or N1MM+) and connect it to kx3, and select the frequency in the log software, will pa100-d switch to the correct band display. Now there are more and more strange problems

The following link is the operation video:
https://drive.google.com/file/d/1egewot ... sp=sharing
Description: The connection in the video is to connect only kx3 and pa100-d. First turn on kx3, then turn on the amplifier. At this time, the amplifier displays "?", then switch the kx3 band, the amplifier still displays "?", then manually switch band+ and band-, and after displaying the band, it becomes "?" again.

[5B4AIY]

You have a serial port problem. First check with a multimeter that your serial port is giving the correct TX DATA voltage. Plug a stereo 3.5mm to 3.5mm patch cable into the PA-100D's RS-232 port, and with a multimeter, measure the output voltage on the TX DATA line, which is the ring connection, it should be around -9V. This verifies that the MAX-232 level converter's charge pump is at least functional.

There is a serial port check feature built into the firmware which allows you to run a loop-back test on the serial port. With the Serial Port set to ON, save the configuration, power OFF. Connect the PA-100D serial port to your computer with a USB-RS232 converter and run the terminal program. Press and hold the DISPLAY/CONFIG button, and power the amplifier on. Release the DISPLAY/CONFIG button when the display indicates the RS-232 loop-back test is running. Typing characters into the terminal program should result in the same characters being echo'ed back. This verifies that the serial port is OK. Press the Power button to cancel the loo-back test.

Annex-A of the User Manual describes how to use the Test facility. I am attaching a ZIP file of a simple terminal program that I have used for these tests. This will also allow you the check the functionality of the serial port.

Once you have verified that your serial port is working you next have to check that when connected to your KX3 it responds correctly. Obviously you will need to set the port speeds appropriately! You can check that the PA-100D responds correctly to the returns from the KX3 by invoking the test identified as Note 3 on page 34 of the User Manual. With the amplifier's serial port still connected to the computer, run the loop-back test. With your port's operation verified, and the amplifier's response to a FA00014175000; data command correct, you can now check your KX3.

Connect your KX3's serial port to your computer using the USB-RS232 converter, run the terminal program, and enter FA; the KX3 should respond with something like FA00014175000; indicating the frequency of VFO-A.

If you use SimpleTerm, launch the program, click on Settings, click on the Base Settings tab to set the com port number and speed, ignore the flow control tab, click on the Device Control tab and select NO. Then click on OK, and finally CONNECT.

These checks should identify where the serial communications problem is located. I am assuming that to port speeds are correctly set, and that you have made sure that the communications parameters are correct.

Adrian, 5B4AIY