PA-100D Bias adjustment is only 300ma

General discussion

PA-100D Bias adjustment is only 300ma

Postby KH2BR » 12 Apr 2012 06:17

My new amp is working ok except for the bias adjustment. The manual calls for 500ma but all I can get is 300ma. Is this ok? If not, what should I look for?

Robert KH2BR
KH2BR
 
Posts: 32
Joined: 13 Nov 2011 07:01
Location: USA

Re: PA-100D Bias adjustment is only 300ma

Postby 5B4AIY » 27 Apr 2012 14:48

Hi, Robert,
Perhaps these notes will help. Let's examine how the bias adjustment for the PA-100D works. If you print out the schematic for this, titled: "JUMA PA100 100W LINEAR AMPLIFIER RF AMPLIFIER BOARD SCHEMATICS", then you can see what's happening. It is important to realise that the PA transistors require a temperature compensated bias supply. You will note from the schematic that Q1 and Q2 are biased by injecting current into their bases from a constant voltage source. If this source were not compensated, then as the transistor's temperature increased, their base-emitter voltage will fall, which would lead to a further increase in collector current, a further rise in temperature, a further reduction of base-emitter voltage, in other words, thermal runaway. To prevent this the voltage source is temperature compensated so that it also reduces as the heat sink temperature rises by the same amount as the base emitter voltage falls.

This is achieved by means of Q3, the BD139 transistor used as both a temperature sensor, and the reference voltage for the bias supply. It is biased by means of a constant voltage source formed by the zener regulator Z1. This means that there is essentially a constant current into the base of Q3. It is mounted on the heatsink, and so as the heatsink temperature increases, the base-emitter voltage of Q3 falls at the same rate as the base emitter voltage of Q1/Q2, approximately -2mV/Degree C. Q4 is a FET used as a switch. When its gate voltage is +5V which occurs when the TX signal is asserted, it conducts, and the bias voltage from Q3 is fed to the non-inverting input of op-amp A1-A. The amplifier is configured as a voltage follower with the current source Q5 and its emitter resistor R21 inside the feedback loop so that the forward gain of the op-amp cancels the base-emitter voltage of Q5 thus ensuring that the voltage across C20 effectively follows that at the output of the bias adjust potentiometer R18.

So much for how it works, so, what can go wrong? First, check that the voltage across Z1 is indeed 5.0V to within a few percent. anything less than about 4.5V or higher than about 5.5 volts would be immediately suspicious. If this voltage is correct, you can check that Q3 is working by displaying the temperature on the front panel. Let the amplifier acquire room temperature for an hour or so, and display the temperature. If you can adjust R14, the offset adjustment pot, so that the temperature reads correctly, and R14 is sensibly near the centre of its travel, not hard at one end or the other, then this shows that Q3 is conducting, and amplifier A1-B is working correctly. Next, with an accurate digital multimeter measure the voltage at the slider of R18, it should be in the range of about 600mV - 800mV, and compare it with the voltage across R20 when in the transmit mode. If Q4 is working correctly, then the voltages should be the same within a few tens of mV. Now measure the voltage across C20. It too should be essentially the same. If you can only achieve a TOTAL bias current of about 300mA, then it means that there is not enough voltage being fed to the bases of Q1/Q2, or that there is a low resistance path there. Check the value of R4, is it 100 ohms? Also check that Q5 is conducting. The voltage on pin 1 of A1-A should be about 600mV - 700mV higher than the voltage across C20 showing that the amplifier is cancelling the forward base emitter voltage of Q5. Measure the base voltage of each transistor, they should be the same.

Another possibility if the bias circuitry is working is that one of your output transistors is not conducting. This is really hard to diagnose because the resistances are so low in this area. You would really have to use an oscilloscope and examine the collector waveforms when the amplifier is being driven. You might also suspect this if the maximum output power was low and you require to remove all the input attenuation. However, let's not get carried away here, the output transistors are quite rugged, not susceptible to static damage, and provided you have not grossly over-driven them are unliklely to fail. They are VERY conservatively rated for this application, check their data sheet, and you will see that each one is capable of supplying 100W PEP, they are rated at 45V, 25A, and 250W collector dissipation, so you would really have to work hard to damage them!

Regards,
Adrian, 5B4AIY
5B4AIY
 
Posts: 214
Joined: 13 Nov 2011 09:22
Location: Cyprus

Re: PA-100D Bias adjustment is only 300ma

Postby KH2BR » 27 Apr 2012 17:09

Hi Adrian,

Thank you for such a nice detailed description of the operation of the bias circuitry. I am so greatfull for your effort.
I will be looking into the problem some time today. I am sure that this analysis will help many others also.

73's, Robert KH2BR
KH2BR
 
Posts: 32
Joined: 13 Nov 2011 07:01
Location: USA


Return to Juma Discussion

Who is online

Users browsing this forum: No registered users and 25 guests