Mark V FP-29


SAFETY NOTE: Troubleshooting this unit should NEVER be attempted without an isolation transformer – signal grounds exist that are 165VDC above AC line neutral. Other voltages exist on exposed parts that exceed 300VDC. EXTREME CAUTION should be used, and only by EXPERIENCED troubleshooters.

I recently received an FP-29 for repair. The supply had no output (s) – no blown fuses – no evidence of smoke – no visible indication of failed components. Application of 120 VAC resulted in a inrush of current, which settled down to less than 2 watts of power consumption.

Cold checking with an ESR meter, a DC ohmmeter, and a Huntron Tracker, revealed:
No bad electrolytic capacitors.
No shorts in Q2 or Q3.
Q4 shorted.
D6 shorted.
D7 open.

These components supply the startup source for the control module HIC-1.

Since no expected voltages were supplied on the schematic, a bit of reverse engineering had to be employed. The defective components would charge C18 to some voltage that would cause the control module to drive Q2 and Q3. A datasheet was obtained for IC-1. This is a device that senses about 4.7 volts, and changes state. R16 and R17
form a voltage divider to supply the input to that device. This established that the voltage expected across C18 was about 13 VDC.

There is no datasheet available for the MB2013A control module. This is a proprietary 12 pin, epoxy coated assembly, mounted vertically on the board.

Since D7 was open – it was a simple matter to apply an external supply across C18, with no AC supplied to the unit.

A scope revealed that at 17VDC, drive pulses appeared at the gate of Q3 – reducing the voltage to 13 volts removed the drive pulses. No drive appeared at the gate of Q2.

Hope still existed that HIC-1 may be OK – since Q2 an Q3 are in a totem pole configuration, and with no schematic of HIC-1 available, it was possible that DC might be required at pin #1 of HIC-1 to cause drive to Q2.

The defective components in the start circuit were replaced.

An isolation transformer, Variac, and wattmeter were used to bring the voltage up slowly while observing the results.

The voltage across C18 rose to about 11VDC at 50VAC input. As the input voltage was increased, the voltage across C18 dropped to about 9VDC.

No output was obtained from the supply, and no drive appeared on Ether Q2 or Q3.

D7 was now removed – with an input of 120VAC, current limited by a 40 watt lamp — an external supply was fed to pin #12 on the control module – and again drive was noted only to the gate of Q3, and no output from the supply. The control module drew about 250MA and became very hot.

At this point an interesting note was made – the voltage at the anode of D6 was about 160VDC – much higher than expected, since the capacitor that it charges (C18) is rated at only 35VDC. This will become very important as we look at this design. In simplistic terms – D11 rectifies AC line voltage – Q4 limits the current – C18 charges towards 15 volts – IC-1 lifts the short on Pin#10 of the control module – the control module drives the FET’s – D9 provides the DC to sustain the control module – D10 causes Q5 to shut Q4 down – D12 causes Q1 to short the soft-start resistors R2 and R3 -the secondary side regulators take over, and normal operation ensues. Output voltage and current information is fed back to the control module by a pair of opto-isolators – PC1 and PC2. These control the control module via pins # 8, 9, 11, and 12.

This explanation was derived from observing the supply after it was repaired. Repair involved replacing the VERY expensive MB2013A control module as well as the earlier mentioned components. C18 was also replaced – even though it checked OK – it may have been overstressed – Q5 was replaced for the same reason.

A POSSIBLE EXPLANATION OF THIS FAILURE : The supply failed to start – Q4 continued to raise the voltage across C18 – the MB2013A module failed – that failure took out the other components.

MODIFICATION TO REDUCE THE POSSIBILITY OF THIS RECURRING: An 18 volt, 1 watt Zener was installed from the Anode of D6, to the negative lead of C18. The Zener diodes Anode towards the negative lead of C18. In the event of any failure of the supply to start, R20, R21, and the added Zener diode should limit the voltage applied to sensitive components.

NOTES ON THE MB2013A CONTROL MODULE: This unit is a proprietary device. Yaesu quoted well over $100 for a replacement. Research developed the fact that it is used in a supply of a different manufacturer. It was obtained for about 33% of the quoted Yaesu price. Comparing the schematics of the two supplies, they are identical up to the transformer, leading us to the conclusion that they were supplied by a vendor. The unknown vendor might possibly be a source of the MB2013A control module. With the exception of this module, and the transformer, all other components are readily
available. The original module was marked MB2013A-1 and the replacement was marked MB2013A-2. Any difference in the later version is unknown.

OTHER NOTES: It appears there are multiple versions of this supply. Some are switchable 120 – 240, others are 120 input only. Some have a thermostatic switch to shut the supply down by shorting pins # 11 and 12 on the control chip, others do not show that. Some supplies have a connector CN-2 some do not. Several components appear in one version, and are deleted in others. The 13VDC output is derived from the 30VDC by a separate regulator – any failure of the main supply
causes loss of 13VDC as well.