LED Diagnosis/Troubleshooting

In order not to use a negative voltage in a MOSFET AGC controlled amplifier, the operating potential of the MOSFET has to be raised 1.5 volts. In a simple 455kHz IF receiver in the ARRL Handbook, the author used an LED as a 1.5 volt zener so that the AGC voltage could run from 0 to 6 volts (instead of -1.5 to 4 volts).

I tried this circuit and found out that not only did the LED light, it's intensity varied with the drain/source current. It also proved to be very useful as an indicator of proper circuit function. When the receiver is finished, a dead amplifier can be quickly located and diagnosed.

Because of the easy diagnosis, the LED's were left in all the amplifiers, regardless of whether they were AGC controlled or not. Gate 2 bias voltage was raised to 6 volts using two 100k resistors.

Over the course of experimentation, the following diagnosis procedure evolved:

LED Does Not Light

>12 Volts is not connected.

>Wrong value decoupling resistor - usually too high (at 12 Volt connection).

>MOSFET installed backwards (In most cases, does not blow the MOSFET).

>Output coil open, installed improperly, or a bad solder joint.

Dim LED - No Output

>Forgot to install input impedance resistor at Gate 1 - Either a 2.2K or a 100K from Gate 1 to ground in this receiver..

>Gate 2 bias voltage too low. Check for wrong resistor from Gate 2 to ground, or bad/missed solder joint on the resistor from Gate 2 to 12 Volts.

Remove 12 Volts to Transformer, Power On

>LED does not light up, good MOSFET.

>LED still lights up - bad MOSFET - Gate 2 shorted to drain.

>To check Gate 1 - short Gate 1 & 2, if LED lights up, bad Gate 1.

>Another quick test is to use a VOM and check the resistance between Gate 1 and Gate 2. If it is very high, the Gates are OK, if the VOM indicates a short, the MOSFET is bad.

Disconnecting the 12 Volts from the transformer is the best way to accurately test for a good MOSFET.

I came across this accidentally, of course, when I was troubleshooting a MOSFET receiver that wasn't working properly, and pulled a transformer out (they were 6 pin MCL T1-1T's) and the LED stayed on. The LED was supposed to go out!

I pulled the other transformers, some had a dim LED, and in others, the LED turned off. In every case where the LED was dim or very bright, I had a bad MOSFET or a solder bridge.

Both MOSFETs in the single balanced mixers can be tested by disconnecting the 12 Volts to the output transformer (with power applied). I found one MOSFET in a mixer that was bad, two others in amplifiers, and one solder bridge between Gate 1 and the Source. The receiver worked great after this diagnosis.

A bright LED indicates proper circuit function. The only time brightness varies is in the AGC controlled amplifiers. The current drawn through the LEDs is approximately 5 mA.

If you want to test for proper brightness, take one of the LEDs, apply 12 Volts through a 2K to 2.4K resistor; the brightness of that LED should be the same as at the amplifiers and mixers.

Back to Top

>If the reverse polarity is with a battery, one or more of the 10 mfd tantalum three pin capacitors will explode. They will be easy to find, the tops will be missing.

The capacitors can be replaced with regular electrolytics. Any value from 10 to 100 mfd will work, whatever can fit in the footprint. Center hole is positive, both outside holes are ground in the footprint.

>Check the 250uH chokes for resistance. If the value is below is below 6 ohms, the choke is bad. Some will still pass current and test good, but if the resistance is 3 ohms or less the wires have shorted and they no longer choke.

Check the chokes at the 12 volt inputs to both boards, the 12 volt feed to the relays half way down the left side of the VFO section, and at the Frequency Counter relay switching.

>Check the diode at the VFO next to the 10.545 relay. This diode may be shorted. It will blow the VN0106N3 at the Phototransistor (between the VFO amps on the right) if shorted and not checked.

>Check all the 5 volt regulators. Sometimes they will crack or blow apart. The 5 volt regulators are at the VFO, BFO, and the Frequency Counter. In most cases only one will blow. The others will be OK.

All the parts past the regulators have not been damaged during the times I have reversed the polarity. It is a good idea on the bad ones to replace the 2.2mfd electrolytics and on the Frequency Counter (if that regulator blew) the 10 mfd electrolytic following the 78L05 regulator.

>Look for burned traces, especially between the 12 volt input connection on Board 1 going to the 12 volt traces on Section 2 and Section 3. Dark LEDs will point to the burned out traces.

>Check the CA3140 on the stabilizer board. It takes 12 volts and is not protected by a regulator. Check for approximately 5 volts output at the relay box.

>If the LEDs do not switch properly at the Bandpass Filters or the Crystal Filters, change the VN0106N3s in the switching circuits next to the filters in pairs. It is difficult to determine which one is bad while in the circuit. Also, the VN0106N3s tend to fail later if not replaced.

>If the LED at the 10.455 relay will not light, the VN0106N3 at the phototransistor, between the first and second VFO amplifiers, will be bad. Voltage will be below 9 volts at the source.

The VN0106N3s are MESFETs. Any MESFET or MOSFET transistor can replace this part. Just be sure to get the pin identifications correct as they vary between part numbers. I will send you replacement VN0106N3s if you email and let me know.

>If there is no VFO output, replace the 2N5486 in the VFO.

>None of the MOSFETs in the amplifiers or mixers have blown with reverse polarity. If one did, the LED in that circuit will let you know.