2m Amp repair KLM

[kj7tsm]

Greetings,
I acquired a KLM PA10-170BL amp, all mode linear. Its output on FM is clean but giving distortion on SSB. We've done a lot of tests with other xcvrs and nailed it down to the KLM. Anyone with experience fixing these Amps or know of someone?
thanks and 73's David KJ7TSM

 

[prcguy]

You can do a few tests yourself like check the bias voltage on the base of the transistors on the regulator side of any choke or RF bead going to the base. If you look up the transistor part # it may give you some info on what bias voltage to expect for class AB operation and that voltage should be the same on both output transistors. You can do the same for any driver transistors. The bias is usually supplied through extra contacts on the TR relay and relay problems can affect the bias.

Does it make rated power in FM mode? If so the transistors are probably ok and it may just be a bias problem. If it doesn't make rated power you can make a simple RF probed and check levels on each transistor base and collector to see if the signal is bigger on the collector and on the pair of output transistors if its the same levels on both.

 

[kj7tsm]

Hello prcguy.
Will check the voltages tomorrow. Does put out full rated power on FM. There are no datasheets for the Motorola SRF 2356 but that is the interchange part for the MRF 245, which was used in the earlier model PA10-160BL. The board in the amp actually says 160BL. I've attached that datasheet. Would the tab to measure voltage be at 6 o'clock?
thanks and 73s David KJ7TSM

 

[kj7tsm]

Hello prcguy.
Will check the voltages tomorrow. Does put out full rated power on FM. There are no datasheets for the Motorola SRF 2356 but that is the interchange part for the MRF 245, which was used in the earlier model PA10-160BL. The board in the amp actually says 160BL. I've attached that datasheet. Would the tab to measure voltage be at 6 o'clock?
thanks and 73s David KJ7TSM

 

[prcguy]

Aah, its a FET. Typically the drain or output (with 12V on it) will be on the Motorola logo side or on the left side of your picture and the drain with the bias will be on the 926 side or the far right. The component on the drain with the green and brown stripe should be an RF choke and I would measure bias on the side of the choke away from the transistor out of view on the picture. I would measure it with no RF drive and it should be close to zero and with RF applied and the relays kicked in maybe 2 volts or so. Whatever it is should be identical on both output transistors. If the voltage is low and does not change between keyed and unkeyed that might indicate a bias supply problem or a relay contact that carries the bias voltage.

The component or resistor with a brown stripe showing coming off the drain to the far left should connect to the other output transistor drain and if so that is for RF power balancing between devices and the stripline power divider at the output.

 

[merlin]

I will check the schemas, but the effects are like all the cheaper "linear" amplifiers. They are NOT linear, they are class C biased means they do OK on AM, FM, or CW, but distort with single sideband.
OK, this amp does look to have a crude bias supply, so no doubt it is out of adjustment or failed.
These designes are a PITA to adjust, you can measure the bias voltage, but it is the IDLE CURRENT that has to be set.
Problem is you need to fire RY1 with no RF drive, then adjust R3 until total current is about 300 Ma. Gate voltage maybe about 4 volts.
No datasheet for the devices, this is just a close guess. (based on MRF245)

 

[prcguy]

I will check the schemas, but the effects are like all the cheaper "linear" amplifiers. They are NOT linear, they are class C biased means they do OK on AM, FM, or CW, but distort with single sideband.
OK, this amp does look to have a crude bias supply, so no doubt it is out of adjustment or failed.
These designes are a PITA to adjust, you can measure the bias voltage, but it is the IDLE CURRENT that has to be set.
Problem is you need to fire RY1 with no RF drive, then adjust R3 until total current is about 300 Ma. Gate voltage maybe about 4 volts.
No datasheet for the devices, this is just a close guess. (based on MRF245)

Yup, it can be difficult to adjust the idle current if there are no removable current measuring points for each transistor collector or drain. Some instructions will have you turn the bias off, place the amp in transmit mode manually with no RF, measure quiescent current then turn up the bias control to draw a specified amount of current. That's easy on a single device amp but more complex with paralleled transistors. I suppose you could adjust one transistor for the recommended current then measure the bias voltage and adjust the other transistor for the same bias voltage.

 

[kj7tsm]

Yup, it can be difficult to adjust the idle current if there are no removable current measuring points for each transistor collector or drain. Some instructions will have you turn the bias off, place the amp in transmit mode manually with no RF, measure quiescent current then turn up the bias control to draw a specified amount of current. That's easy on a single device amp but more complex with paralleled transistors. I suppose you could adjust one transistor for the recommended current then measure the bias voltage and adjust the other transistor for the same bias voltage.

Thanks. I'm getting some local help to do this testing. Both of you have been very helpful. 73s David KJ7TSM

 

[kj7tsm]

Hello gentlemen
A couple of retired engineers in our local club took it on as a project. Here is their summary:

Here is what the amp measured before I made any changes:

Driver Base .683V
Final Base .422V

Current with both diodes shorted: 560mA
Current with final diode shorted: 580mA
Current with no diodes shorted: 580mA

Here is best I could do by changing bias to where it leveled off:

Driver Base .701V
Final Base .682V
Current with both diodes shorted: 560mA
Current with final diode shorted: 600mA
Current with no diodes shorted: 710mA

Output is 160W with 10W in FM

The results from signal reports show about 95% reduction in distortion. The amp is perfectly usable.
I am very pleased with the result.
73s David KJ7TSM