Pioneer SX-650 Receiver Restoration

Unit: AM/FM Stereo Receiver

Manufacturer: Pioneer

Model: SX-650

SN: WL3633228S

Today I'm showcasing a Pioneer SX-650 receiver that came in for restoration. The SX-650 was manufactured from 1976 to 1978. It produces 35 watts per channel into 8 ohms with no more than 0.3% total harmonic distortion. The list price in 1976 was around $300. This particular unit came in with several faulty components. However, the receiver is still in excellent cosmetic condition. The wood case, faceplate, and all knobs are almost flawless. It is certainly worth restoring. According to the Pioneer database, this unit was manufactured in December 1976.

Initial Troubleshooting

The previous owner told me that one channel is highly distorted and the other channel sounds very weak. First of all, I decided to check the operating voltages on Sanken SP-40W Power Modules to make sure that nothing was wrong with them. There is no modern substitution for those modules and the new modules available on eBay are usually counterfeit from China. A quick test showed that all measured voltages on both modules were very close to the reference voltages on the schematic. Well, at least these power modules are not burned out.

Test results on Sanken SP-40W power modules:

pin 1: schematic: 36V, left channel: 36.2V, right channel: 36.1V

pin 2: schematic: 1.2V, left channel: 1.3V, right channel: 1.3V

pin 3: schematic: 65mV, left channel: 63mV, right channel: 58mV

pin 4: schematic: 33mV, left channel: 37mV, right channel: 31mV

pin 5: schematic: -36V, left channel: -36.1V, right channel: -36.1V

pin 6: schematic: -1.2V, left channel: -1.2V, right channel: -1.2V

pin 7: schematic: -6mV, left channel: 9.8mV, right channel: 6.2mV

pin 8: schematic: N/A, left channel: 0V, right channel: 0V

The next test I completed was to trace the input signal through the balance/volume control, tone control, power amplifier - 1, power amplifier - 2, and down to the speaker's terminals. This test will show me where the signal is distorted or attenuated. So, I connected my function generator to the AUX terminals, applied a sine-wave signal of 1 kHz, and traced the input signal with my oscilloscope throughout all boards down to the speaker's terminals. Surprisingly but the sine-wave signal was clear and without any distortion throughout all boards. I connected my cell phone to the AUX terminals and the receiver played music with a crystal clear sound. It looks like the amplifier in this unit has no issues.

So, the last step was to check the tuner performance. I connected the FM antenna to 300 Ohm terminals of SX-650 and switched to FM broadcasts. Tuned the knob to the local FM station and immediately realized that the left channel was very weak and the right channel was highly distorted. At the same time, the signal strength was very strong and the stereo indicator lamp came on. Well, there is definitely some issue in the tuner circuit.

I turned off the receiver and tested all transistors and diodes in the tuner circuit. All semiconductor devices were normal. Then, I turned on the receiver again and checked the operating voltages on all transistors and integrated circuits (ICs). The measured voltages on all six transistors Q1 thru Q4, Q7, and Q8 were very close to the reference voltages on the schematic. Two ICs Q5 and Q9 also had the correct operating voltages on all pins. However, the integrated circuit Q6 was in question. The measured voltage on pins 4, 6, and 7 didn't match the reference voltage on the schematic.

Test results on integrated circuit Q6:

pin 1: schematic: 14.0V, measured: 14.0V

pin 2: schematic: 3.0V, measured: 3.1V

pin 3: schematic: 7.5V, measured: 7.7V

pin 4: schematic: 11.0V, measured: 4.8V

pin 5: schematic: 11.0V, measured: 11.2V

pin 6: schematic: 7.3V, measured: 9.5V

pin 7: schematic: 7.3V, measured: 4.9V

pin 8: schematic: ground, measured: 0V

pin 9: schematic: 0.1V, measured: 1.0V

pin 10: schematic: 2.3V, measured: 2.5V

pin 11: schematic: N/A, measured: 2.4V

pin 12: schematic: N/A, measured: 0V

pin 13: schematic: 2.2V, measured: 2.5V

pin 14: schematic: 2.4V, measured: 2.5V

pin 15: schematic: 2.4V, measured: 2.5V

pin 16: schematic: 3.0V, measured: 3.2V

The integrated circuit Q6 (HA1196) installed on Pioneer SX-650 is an MPX decoder. Pins 4 and 5 are the left and right channel outputs, respectively. And pins 6 and 7 provide feedback control. The sound from the left channel is very weak when the receiver is switched to FM broadcasts. It makes sense since the measured voltages on both pins (4 and 7) don't match with the reference voltages on the schematic. The right channel has a sound but it is highly distorted. It also makes sense since the measured voltage on pin 5 is close to the reference voltage but there is a problem with feedback control on pin 6. So, it looks like a faulty IC. I unsoldered the IC Q6 from the board and installed a new HA1196 chip on the 16-pin DIP socket. Then, I turned on the receiver, tuned the knob to the local FM station, and... The left channel was still very weak and the right channel was still highly distorted. Huh! It looks like there is another faulty component on the tuner board. I carefully inspected all passive electronic components around the HA1196 chip and found that the polystyrene film capacitor C41 (510pF) is faulty. I replaced it with a new polystyrene cap and there you go! The clear sound from FM broadcasting was restored. No more distortions. The measured voltages on all pins match with the reference voltages on the schematic. Yep, the troubleshooting of the tuner circuit is always a tricky task. Now it's time to service other boards.

Test results on integrated circuit Q6 after HA1196 chip and capacitor C41 replacement:

pin 1: schematic: 14.0V, measured: 14.0V

pin 2: schematic: 3.0V, measured: 3.1V

pin 3: schematic: 7.5V, measured: 7.7V

pin 4: schematic: 11.0V, measured: 11.1V

pin 5: schematic: 11.0V, measured: 11.1V

pin 6: schematic: 7.3V, measured: 6.7V

pin 7: schematic: 7.3V, measured: 6.7V

pin 8: schematic: ground, measured: 0V

pin 9: schematic: 0.1V, measured: 1.0V

pin 10: schematic: 2.3V, measured: 2.5V

pin 11: schematic: N/A, measured: 2.4V

pin 12: schematic: N/A, measured: 0V

pin 13: schematic: 2.2V, measured: 2.3V

pin 14: schematic: 2.4V, measured: 2.5V

pin 15: schematic: 2.4V, measured: 2.5V

pin 16: schematic: 3.0V, measured: 3.3V

Old and new integrated circuit Q6 (HA1196)

16 pin DIP socket for integrated circuit Q6

New HA1196 chip installed on the socket

Power Amp - 1 Board

The power amp - 1 board (#GWH 101) has two tantalum capacitors C1 and C2 installed in the signal path. I replaced them with high-quality film polyester WIMA MKS2 caps. Other 6 e-caps C7 thru C10, C13, and C14 were replaced with low impedance and high-reliability Nichicon UPW caps. The protection relay S1 installed on this board is very clean inside and I didn't notice any contact degradation. So, I didn't replace it. The modern substitution for this relay is Omron MY4-02-DC24.

Power Amp - 1 (#GWH 101) - before and after

Power Amp - 2 Board

The original coupling capacitors C21 and C22 installed on the power amp - 2 board (#GWR 101) are sky blue Sanyo electrolytic caps. These e-caps become very leaky over time and should be always replaced in any vintage gear. For more information about sky blue Sanyo e-caps refer to my previous post on Pioneer SX-828 restoration. I replaced them with high-quality film polyester Kemet caps. The other 15 e-caps installed on this board were replaced with new Nichicon UPW/UPM caps. I also removed and tested two transistors Q7 and Q8 mounted on the heat sinks with Atlas DCA55 semiconductor analyzer. Both transistors passed the test and the measured DC current gain was in spec according to the datasheet. I applied a new silicone thermal compound between each transistor and the heat sink before installation on the board.

Sky blue Sanyo coupling capacitors C21 and C22 - replaced with film polyester Kemet caps

Power Amp - 2 (#GWR 101) - before and after

Transistors Q7 and Q8 were removed from the board and tested with Atlas DCA55 semiconductor analyzer

The old thermal grease is almost completely dried out

Tone Control Board

The tone control board (#AWG049) has only 8 electrolytic capacitors but this board is not easily accessible to work on. I would strongly recommend removing the faceplate and the dial frame before servicing this board. I don't think it would be possible to install any new electronic components without removing the faceplate and the dial frame. Even when these two parts are removed it is still a bit tricky to install some particular capacitors.

This board has two tantalum capacitors C1 and C2 which were replaced with high-quality film polyester WIMA MKS2 caps. Two low leakage e-caps C13 and C14 were replaced with new low leakage Nichicon UKL caps. The remaining four electrolytic capacitors were replaced with low impedance and high-reliability Nichicon UPW caps.

Faceplate and dial frame removed - Tone control board is behind the pots

Tone control board - before and after - very limited space to work on

Tuner and AF Board

On the board (#AWE075) I replaced all tantalum capacitors C43 thru C50 installed in the signal path after the multiplex decoder circuit. The capacitors C43 and C44 were replaced with low leakage Nichicon UKL caps. Four capacitors C45 thru C48 were replaced with high-quality film polyester Kemet R82 series, and the last two C49 and C50 with film polyester WIMA MKS2 caps.

The phono equalizer and microphone circuits are both a part of the tuner and AF board. The phono equalizer has 8 e-caps: C76, C77, C84, C85, C90, C91, C94, and C95. Four of them C76, C77, C90, and C91 are tantalum caps. I replaced them with low leakage Nichicon UKL caps. The remaining 4 caps were replaced with low impedance and high-reliability Nichicon UPW caps. 

The microphone amp has four e-caps: C100, C101, C104, and C105. The e-caps C100 and C104 are tantalum caps and were replaced with high-quality film polyester Kemet R82 and WIMA MKS2 caps, respectively. The remaining 2 caps were replaced with Nichicon UPW caps.

Tuner and AF board - before and after

Phono equalizer circuit - before and after

Microphone amp circuit - before and after - very limited space to work on

Output Power Test

At the end of my restoration, I loaded this receiver with a low inductance 8Ω/100W dummy resistor for each channel, connected my oscilloscope across the speaker terminals, and applied a sine-wave signal of 1 kHz to the AUX jacks. The output sine-wave signal was perfectly symmetrical on both channels with no clipping up to 15.73 VRMS (left channel) and 16.63 VRMS (right channel). So, one can calculate that the output power is (15.73x15.73)/8=31W (left channel) and (16.63x16.63)/8=35W (right channel). It is very close to factory specifications. A very small difference between the left and right channels is attributed to the imperfection/aging of the balance potentiometer. Keep in mind that this baby is almost 43 years old but still in perfect working condition!

Output power test

As usual, all the knobs and the front panel were gently cleaned in warm water with dish soap. All knobs were also slightly polished by Mothers Mag & Aluminum polish to remove some small spots of aluminum oxidation. All pots and switches have been cleaned with DeoxIT 5% contact cleaner and lubricated with DeoxIT FaderLube 5% spray.

The final result is amazing. The receiver looks brand new. The sound is wonderful and it has plenty of power for home stereo needs. Please watch a short demo video at the end of this post. Thank you for reading.

Pioneer SX-650 - before restoration

Pioneer SX-650 - after restoration

Demo video after repair & restoration

Pioneer SX-780 Receiver Restoration

Unit: AM/FM Stereo Receiver

Manufacturer: Pioneer

Model: SX-780

SN: ZI3608397Y

I restored this receiver about a year ago but unfortunately didn't take a lot of "as is" photos. However, I thought it would still make sense to write a post about my restoration process on this beautiful classic receiver. 

The Pioneer SX-780 is a great mid-range receiver from one of Pioneer’s last series of receivers before they switched to digital meters and black faceplates. It features a "high output, low distortion power amplifier, a high fidelity equalizer amplifier with low noise and high gain, a high stability tuner section, tone controls with tone defeat function, built-in protection circuits and independent power meters". The Pioneer SX-780 was manufactured from 1978 to around 1980. It was initially built in Japan but later production switched to South Korea as evidenced by the varying tags on the back of the unit. It produces 45 watts per channel into 8 ohms with no more than 0.05% total harmonic distortion. The list price in 1978 was $325.00. According to the Pioneer database, this unit was manufactured in September 1979.

I bought this receiver from the second owner and it had all the typical issues associated with dirty pots and switches. The right channel was very scratchy and the left channel had no sound at all. I was afraid that the Darlington power pack from the left channel was dead but fortunately, it was just a VERY dirty volume pot. The receiver was serviced at least once in the past but only two filter capacitors C501/C502 were upgraded from 8000uF/50V to 10000uF/50V Nichicon UKW series. Everything else looks original and untouched on all boards. Below are some photos I took during my restoration process, complete with comments and tips.

AF Amplifier Board GWK-118

Three transistors (Q19, Q20, and especially Q25) in the power supply circuit are running very hot and can even unsolder themselves. Those transistors are mounted on heat sinks but the size of each heat sink is not enough to efficiently dissipate the heat. I would strongly recommend carefully investigating a PCB around those transistors for any signs of excess heating if you need to service this model. Be aware that those heat sinks are electrically live.

I decided to replace all three transistors with modern Fairchild MJE transistors due to better thermal characteristics in comparison to original transistors. A new silicone thermal compound (Wakefield-Vette, 120 series) was applied between each transistor and the heat sink.

Three transistors Q19, Q20, and Q25 in the power supply run very hot - heat sinks are electrically live!

Three original transistors Q19, Q20, and Q25 were replaced with Fairchild MJE transistors

New Fairchild MJE transistors installed

All electrolytic capacitors in the power supply and protection circuits were replaced with low impedance and high-reliability Nichicon UPW caps. Those e-caps are designed for switching power supplies and can operate in the temperature range from -55 to +105 C. The protection circuit in SX-780 is responsible for muting when the POWER switch is turned ON and OFF. The delaying action is determined by the time constants of the timing capacitor C317 and the constant current circuit. The transistor Q26 is turning the relay S7 on and off thereby protecting the speakers.

I found an error in the schematic concerning the e-cap C311. The schematic shows this capacitor as 100uF/16V despite the original capacitor installed on the board being 220uF/16V. I replaced this e-cap with a Nichicon UPW capacitor, 220uF/25V.

Another issue in the schematic is related to four e-caps C324 thru C327 installed in the power meter circuit. The unit I own was manufactured in South Korea and there are only two capacitors C324 and C327 installed on the PCB. Two jumpers are installed instead of C325 and C326. I believe this is a result of production optimization which was not documented by Pioneer. So, I replaced those two capacitors with new Nichicon UPW caps.

Only two capacitors C324 and C327 installed on the PCB in the power meter circuit

This board has 12 low leakage electrolytic capacitors installed in the signal path (all orange caps). These are C103/C104, C117/C118, C207/C208, and C217 thru C230. I replaced two e-caps C207/C208 with high-quality film polyester WIMA MKS2 caps. The other two e-caps C227/C228 were replaced with film polyester Nichicon caps. And the remaining 8 capacitors were replaced with low leakage Nichicon UKL caps.

Test results on original low leakage capacitors removed from AF Amplifier board:

C103: rated capacitance – 2.2uF, measured – 2.5uF, deviation: +14%

C104: rated capacitance – 2.2uF, measured – 2.5uF, deviation: +14%

C117: rated capacitance – 2.2uF, measured – 2.5uF, deviation: +14%

C118: rated capacitance – 2.2uF, measured – 2.6uF, deviation: +18%

C207: rated capacitance – 1uF, measured – 1.07uF, deviation: +7%

C208: rated capacitance – 1uF, measured – 1.03uF, deviation: +3%

C217: rated capacitance – 4.7uF, measured – 5.7uF, deviation: +21%

C218: rated capacitance – 4.7uF, measured – 5.7uF, deviation: +21%

C227: rated capacitance – 0.22uF, measured – 0.25uF, deviation: +14%

C228: rated capacitance – 0.22uF, measured – 0.23uF, deviation: +5%

C229: rated capacitance – 2.2uF, measured – 2.4uF, deviation: +9%

C230: rated capacitance – 2.2uF, measured – 2.4uF, deviation: +9%

All other electrolytic capacitors on this board were replaced with low impedance and high-reliability Nichicon UPW caps. I found one extra e-cap on this board that was not marked on the circuit board schematic. It is just to the right of e-cap C309 and the value is 4.7uF/35V. It looks like this is also a result of production improvement which was not documented by Pioneer. I replaced this cap with a new Nichicon UPW.

Extra e-cap which is not marked on the circuit board schematic - 4.7uF/35V (right of e-cap C309)

New Nichicon UKW series filter capacitors C501 & C502 - upgraded by the previous owner

Tuner Board AWE-099

The tuner board has two transistors Q7 (2SA726) and Q8 (2SA726) which are prone to create noise. These transistors are gain matched (blue paint dots on the top of each). I replaced them with modern low noise Fairchild KSA992. The new transistors were gain matched within 1%. I didn't change any capacitors on the tuner board.

Tuner board - transistors Q7 and Q8

Darlington Power Pack

As I mentioned earlier this receiver came in with a very scratchy right channel and the left channel had no sound at all. The first suspect was a faulty STK-0050 Power Pack which is a fairly common issue in this model. Any speaker shorting results in excessive current and can easily blow these power packs. So, I checked the operating voltages on all pins and compared them against the schematic. The voltages on all pins were very close to those on the schematic. This was a good sign meaning that both STK-0050's packs are not defective. Indeed, after a deep and intensive cleaning of all switches and pots with DeoxIT the sound in both channels was restored. Yep, sometimes those old gears require just a very good cleaning and lubrication. I carefully unsoldered both Darlington packs, removed the old thermal paste, and applied a new silicone thermal compound between each pack and the heat sink. It is always a good idea to refresh the thermal paste to improve the heat dissipation on any output devices.

Original STK-0050 power packs

STK-0050 power pack removed from the PCB

Dial Lamps

Initially, I installed modern warm white LED lamps instead of old incandescent bulbs but they look too bright on this model. So, I moved to new incandescent bulbs. To reflect the light and diffuse the heat coming from these bulbs I installed a strip of foil tape inside the top cover. It substantially reduces the heat coming from incandescent bulbs.

Tip: there is a small opening at the center of each lamp socket. I used a small Allen wrench and gently pushed each lamp out of the socket. In that case, the brittle socket tabs won't be damaged or broken.

Push each dial lamp out of the socket with a small Allen wrench

DC Balance & Power Meters Adjustments

At the end of my restoration, I checked and adjusted the DC balance of the power amplifier as described in the service manual. No dummy load or input signal is required for this adjustment. The DC balance was adjusted to ~7mV on each channel.

An AC voltmeter and function generator are required to adjust power meters. A sine-wave signal of 1 kHz should be applied to the AUX terminals and the level of this signal should be adjusted so that the voltage on the SPEAKERS terminals read 20 VRMS. Then, the trimming resistors VR7 and VR8 should be adjusted so that the power meters read 50W.

DC balance adjusted to ~7mV on each channel after restoration

Output Power Test

The final output power test was conducted at the end of the restoration. Two low inductance 8Ω/100W resistors connected across each speaker terminal were used as a dummy load. The output sine-wave signal was perfectly symmetrical on both channels with no clipping up to 20.51 VRMS (left channel) and 20.75 VRMS (right channel). The output power on each channel can be calculated based on the following formula: P = (VRMS x VRMS)/8. The max output power (before clipping) is 52.6W on the left channel and 53.8W on the right channel. This corresponds to the factory specifications of this model and even slightly exceeds them.

Output power test - perfectly symmetrical output sine-wave signal with no clipping up to ~53W

As usual, all the knobs and the faceplate were gently cleaned in warm water with dish soap. I also slightly polished all knobs by Mothers Mag & Aluminum polish to remove some small spots of aluminum oxidation. All pots have been cleaned with DeoxIT 5% contact cleaner and lubricated with DeoxIT FaderLube 5% spray.

The final result can be seen in the photos below. The sound is wonderful, very warm, and extremely clean. Please watch a short demo video at the end of this post. Thank you for reading.

Pioneer SX-780 - before restoration

Pioneer SX-780 - after restoration

Demo video after repair & restoration