RF Transistors, if they test 0.7V drop can they still be bad? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

RF Transistors, if they test 0.7V drop can they still be bad?

Move Thread

LAN_403

zero_coke

Thu Mar 21 2013, 05:24PM

Registered Member #8368 Joined: Wed Nov 28 2012, 04:11AM
Location:
Posts: 9

We've been recently encountering some problems with our AN762 1.8-18 MHz amplifier. It uses two MRF454 transistors and the amplifier doesn't work.

We removed the transistors and placed a resistor from base to collector/emitter and tested the voltage drop to be 0.7V so meaning the biasing is correct. But when we place the transistors they should technically be on but they're super cooland we get no amplified signal at the output. The funny thing is, when we remove these transistors and test them separately using a diode test we do get 0.7V drop across them so now we don't know whats going on. Can they still be damaged even though they pass the diode test?

Dr. Slack

Thu Mar 21 2013, 07:43PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Yes. It's like kicking the tires of a car to check whether it will go.

zero_coke

Thu Mar 21 2013, 09:08PM

Registered Member #8368 Joined: Wed Nov 28 2012, 04:11AM
Location:
Posts: 9

Can you please elaborate on that. What do you mean? So how can they pass the diode test and still be damaged?

Dr. Slack

Fri Mar 22 2013, 08:53AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

How can they pass the diode test and still be damaged? About as easily as a car can have 4 inflated tires, and a bucket of nails for the engine.

Try this slightly more comprehensive but still simple bias test. R2 and R3 set up a potential divider. If the transistor is working, it should establish 0.7v on the base, so will end up with around 1.4v on the collector. Now connect a 10k from the base to either ground or +5v, and the collector voltage should move.

If it does all those things, then I'd start to be surprised if it didn't bias up and at least do some sort of amplification in the target RF circuit.

johnf

Fri Mar 22 2013, 09:14AM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

Zero
RF transistors are made up of many small HF transistors in parallel
damage to any one will seriously limit the ability to amplify at RF.

PS you should not use a multimeter to measure the base emitter junction on the ohms or diode function --this can cause damage. It is far better to measure voltage in circuit with and without RF applied.
High VSWR is a real killer to bipolar transistors and mosfets but bipolars are more susceptible.
Even a 1dB attenuator on the output will protect them with a 20 % output power loss

Steve Conner

Fri Mar 22 2013, 10:31AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Put the transistors back in the amp, go through the bias setting procedure in AN762 and tell us the results. Can you get the amp to draw the recommended idle current? This should make the heatsink get slightly warm even with no signal, and the bias current should stay reasonably stable as it warms up.

If it doesn't bias up properly, you probably made a mistake while assembling the kit. (The amp is sold as a kit, right?) Don't bother trying to apply RF until you get this sorted out.

If it biases up but won't amplify, I'd still suspect an assembly error. The transformer windings are complicated, and if you got one backwards it could cancel out the signal.

Re Dr. Slack's comment: The diode test function on a multimeter applies a tiny current, roughly 1mA. When you use it to test a power transistor, it's like testing a Porsche by driving it round the parking lot in 1st gear.

A partly damaged power transistor could handle 1mA but be unable to deal with the 20 or even 50 amps of RF required to make the rated output power.

However in this case, I'd expect it to still amplify RF at low power. I'd also expect it to progress quickly from "partly damaged" to "completely dead" when you tried to bring it up to full power, much as your Porsche engine might.

zero_coke

Fri Mar 22 2013, 05:49PM

Registered Member #8368 Joined: Wed Nov 28 2012, 04:11AM
Location:
Posts: 9

Wow thanks guys for all the helpful replies.

@Dr Slack: I will try that comprehensive test as soon as I go back home

@johnf: Thanks for the info!

@Steve: When we have the transistors in the amp, one resistor in particular (R12, the 20 ohm 5Watt) in the attached schematic click here gets really really hot and that's it. The transistors are super cool and heatsink is super cool as well. Only 1 resistor gets hot and its the R12.

We checked every component 1 by 1 and all was soldered correctly and no shorts or bridges anywhere. However, when my partner initially tested this thing for the first time after we assembled it from the kit, he said there was a short between the collector and emitter on the pcb board on one of the RF transistors. Could this initial test with the short there have damaged them? He refuses that idea because he said he removed and tested the transistors (MRF454's) using a multimeter diode test and they seem to give the correct 0.7V drop in the BE and BC junctions.

I still believe the transistors are damaged because all other components are passive and seem ok when I did a thorough inspection. The only other thing I can think of is the 10 pin voltage regulator IC that's on the board which may have been damaged when that short was in place on the transistor. Other candidates are the only other 2 transistors (The 2N6488 pass transistor or the 2N5190 transistor which acts as a diode basically).

Before we applied any small signal input, we tested the bias voltage on the transistors and it shows up correctly as 0.7V (the manual says we're to observe a value 0.5-1V and adjust the potentiometer to get 0.7 and we did that). Then it said to set the collector current of the transistors by placing a multimeter in series with the collector and the Vcc and measure the bias current. When we tried this we had 0 reading in the multimeter.

I just want to mention other things that we've done that might trigger some stuff in your minds:

My partner always plays with the trimpot on the board because he thinks by doing so he'll fix the problem. Also, when we built the amplifier and tested for the first time, he connected the output of the amplifier directly to an oscilloscope to measure the voltage (this is when the transistors also had a short on the pcb board from collector to emitter). Can you guys suspect anything with this behaviour?

I don't know what to do anymore...we ordered replacement parts for the ICs and hopefully this will solve it as there are no other issues it seems with the passive components on the board nor any other board shorts/bridges that we can observe. Heres a picture of the assembled amplifier (this is before we desoldered and soldered the transistors a hundred times)

202

zero_coke

Fri Mar 22 2013, 10:49PM

Registered Member #8368 Joined: Wed Nov 28 2012, 04:11AM
Location:
Posts: 9

Nevermind...I think we forgot to install the L3 and L4 current probes! Jeez, we thought they were optional and we didnt install them. I guess they were critical!

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob