According to my research a bipolar RF power transistor is likely to have several paralleled transistors inside a single case. They each have fused emitter resistors. Some could be blown, but the unit still appear good on a diode checker. But place it in an amp designed for it and it will have low output power. Not sure if a diode checker would damage it unless it applies test voltages exceeding the transistor ratings.
From: Dan Grimes [mailto:email@example.com]
Sent: Friday, November 16, 2018 1:49 PM
Subject: Re: [CRTech] Testing RF Power Transistors
I thought you could check with a diode checker, too. But recent testing has brought me some trouble.
I bought some ASI 2N5643 from Mouser and then proceeded to test them (before I installed them) with a DMM on the diode setting. To my surprise, I got a P-N Junction from the Emitter to the Collector (but open the other direction) which seemed faulty to me. So I contacted Mouser and ASI to see if this is expected. Mouser Tech Support couldn't help but ASI sent me an information sheet with the following statement:
"Dear Valued Customer,
"It is not possible to correctly measure the collector to emitter junction on a bipolar RF transistor with a digital multimeter. Such testing can damage RF power transistors. The problem with using a digital multimeter is the impedance of the meter and the voltage/current source used by a digital multimeter to measure resistance. The correct test for measuring the junction of an RF transistor is break down voltage and leakage current.
"To measure break down voltage (Bvceo, Bvebo, Bvcbo) you need a current source capable of supplying a specified current to the junction. The junction breakdown voltage is the minimum acceptable voltage specified for that wafer process/device type at a specified force current.
"To measure junction leakage current (Ices, Icbo, Ices) you need to force a specified voltage across the junction and measure the current at that specified voltage. The current measured must be below the maximum specified current for that wafer process/device type.
"Because multimeter testing can damage RF power transistors, any such testing will completely void the ASI warranty. We appreciate your cooperation and please contact us if we can provide any further information on this matter.
"Production Department ASI Semiconductor, Inc. "
ASI told me Mouser should have included this my order but Mouser did not.
ASI did not directly tell me that I ruined the transistors or that a P-N junction should not be seen on them but they are leading me to believe that I ruined them and they will not warranty them. I know that if the transistor is in saturation, it can conduct current from Emitter to Collector. But I am not biasing the Base externally. Perhaps it is internally? ASI has not yet responded to my question if this is expected or if it definitely is off.
At $75 each from Mouser, I do not want to simply throw them away if they are good. In the past, I have tested Motorola 2N5643 transistors in the usual way (diode checker) and this has seemed to work fine. But I have never seen a voltage drop either direction on the Collector to Emitter or visa versa. So now I need to find a relatively easy way to test them without further damage. Otherwise, I must install them in the circuit to test.
On 11/16/2018 12:57 PM, Allard, Mark G wrote:
My assumption is that if these are typical NPN/PNP transistors and you are using a digital multi-meter that might even have a diode test mode, you should be safe to test these junctions without insodent.
From: Dan Grimes <firstname.lastname@example.org>
Sent: Friday, November 16, 2018 14:47
Subject: [CRTech] Testing RF Power Transistors
I need to test some 2N5643 and MRF315A power transistors for RF (VHF). I have been told that using a DMM may damage them and they need very specific currents and voltages to test.
Does anyone have a good and simple test bed or method for testing a transistor before they are installed in a circuit?
For CRTech resources visit http://CRTech.org/ To unsubscribe, e-mail: email@example.com List problems? E-mail: TechStaff@CRTech.org