Leestereo's Restoration & Upgrade of a Pioneer SX-750

Leestereo

Super Member
Here are some pictures and details of a Pioneer SX-750 restoration/upgrade project.

Power Supply and Protection Assembly (AWR-099)
This particular example exhibited a constant loud "hum" noise due to the failure of the filtering capacitors (C17, C18) in the regulated power supply (AWR-099); this is a relatively common problem for this model.

PS%20amp%20Protection%20AWR-099%20gray_zpssewhzeag.jpg


The original capacitors for C17 and C18 were rated at 330µF/50V and these were replaced with 470µF/63V low ESR types. Note that it is recommended that the voltage rating for these capacitors be increased from 50V (today's higher AC mains voltage results in these capacitors being subject to >50VDC). Capacitors C10, C15 and C16 were originally rated at 100µF/50V and were replaced with 220µF/50V low ESR types. The capacitors C12, C13 (zener shunts) were originally rated at 220µF/16V and 100µF/50V (respectively) and were replaced with low ESR types rated at 220µF/35V and 100/63V (respectively). The capacitor C11 was originally rated at 4.7µF/35V and was replaced with a 4.7µF/50V low ESR type. The final capacitors in the regulated PS (C7, C8 and C9) were were originally rated at 220µF/16V, 100µF/35V and 100µF/50V (respectively) and were replaced with low ESR types rated at 220µF/35V, 100µF/50V and 100µF/63V (respectively).

The original capacitors C1 and C2 (infamous failure-prone Sanyo solid aluminum type) in the Protection circuit were originally rated at 0.22µF/10V and were replaced with 0.22µF stacked film types. The C3, C4 capacitor pair was originally rated at 330µF/6V and were replaced with 330µF/25V low ESR types. Capacitors C5 and C6 in the relay timing circuit were originally rated at 3.3µF/10V and 220µF/16V and were replaced with low ESR types rated at 3.3µF/50V and 220µF/35V (respectively). The original DEC MS4U relay was replaced with an Omron MY4-02 24V relay. Also a small heatsink was installed on Q12 (2SB507); it runs considerably cooler now.

SX-750%20Power%20Supply%20Restored%20Label_zps7g78ihmo.jpg
 
Power Amplifier Assembly (AWH-046)
This particular SX-750 (manufactured in July 1976) uses a variant of the AWH-046 board which includes a 1.0µF film capacitor by-pass for the the electrolytic capacitor in the feedback loop.

Power%20Amp%20AWH-046_zpsxwljr9w2.jpg


Signal Path:
The original capacitors (C1, C2) in the input high-pass filters were 1.0µF/25V tantalum types; these were upgraded to 1.0µF/63V WIMA stacked film types. Similarly, the garden-variety 120pF ceramic capacitors (C5, C6) in the low-pass filters were upgraded to 47pF C0G types (the lower value eliminates any high frequency phase shift within the audible range due to the filter F3). The capacitors C7 and C8, which set the low end cut-off of the negative feedback loop, were originally 220µF/6V polarized electrolytic types and were replaced with a 100µF/25V Nichicon Muse ES bi-polar audio grade capacitors.

The 2SA798 (2 transistors in single package) was used for the input differential pairs (Q1-a,Q1-b and Q2-a,Q2-b). To correct the high (~80mV) DC offset measured in the left channel, the 2SA798 was replaced with a pair of hFE matched KSA992 transistors. The right channel 2SA798 was also replaced to match the left channel. Note that the AWH-046 board is designed to accept the discrete transistor pairs at the input without any modification.

Non-signal Path Capacitors:
The stock local decoupling capacitors, C9 and C10, were rated at 100µF/35V and these were replaced with 220µF/50V low ESR types. The emitter by-pass capacitors (C14, C15) for Q3 and Q4 (respectively) were originally rated at 220µF/6V; the replacements were 220µF/25 low ESR types.

SX750%20Power%20Amp%20Restored%20label_zpsyspmtapf.jpg
 
Tone Control Amplifier Assembly (AWG-046)

Tone%20Control%20Amp%20AWG-046_zps68wuvgyj.jpg


Signal Path:
The stock capacitors (C1, C2) in the input high-pass filters were rated at 0.47µF/25V and were the infamous failure-prone Sanyo solid aluminum type; replacements were 0.47µF/50V bi-polar types. The capacitors in the feedback loop (C3, C4), were originally 100µF/6V polarized electrolytic types and their replacements were 47µF/25V Nichicon bi-polar types. Note that the increase in the F3 (to 1.3Hz from 0.6Hz) is inconsequential. The capacitors C9 and C10 at the TA7136P output were originally 4.7µF/25V polarized low leakage types; the replacements were 4.7µF/25V Nichicon bi-polar types.

Non-signal Path Capacitors:
The stock local decoupling capacitors, C7 and C8, were rated at 220µF/35V and the replacements were 220µF/50V low ESR types.

SX-750%20preamp%20label_zpsf8p19uuu.jpg
 
Going to see one tomorrow. What is the general opinion regarding the sx750? Between 300 for a sx750 and 200 for a sx550, would you choose?
 
...Between 300 for a sx750 and 200 for a sx550, would you choose?

IMO, the SX-750 is much better than the SX-550; the 750 has more in common with the higher models (e.g. 850 and higher) than the models below it. Note that for $300 I would expect it to be serviced and in excellent overall condition.
 
Phono Stage (AWE-073 & AWK-065)
This particular SX-750 uses a pair of small boards (AWK-065) with discrete transistors (mounted inside the back panel) to replace the Toshiba TA7136P1 IC. Also noteworthy, is the use of 1% metal film resistors (R71-R76) in the RIAA feedback loop rather than common 5% carbon film types. The capacitors (C94-C97) in the RIAA loop are high quality polystyrene types.

Phono%20Stage_zps3ddhgebj.jpg


Signal Path Capacitors:
The original capacitors (C82, C83) in the input high-pass filters were 1.0µF/25V tantalum types; the replacements were 1µF/50V bi-polar types. The DC blocking capacitors in the RIAA feedback loop (C88, C89) were originally 100µF/6V polarized types, these were replaced by 47µF/25V bipolar types. Note that the increase in the F3 (to 2.2Hz from 1.0Hz) is inconsequential. The capacitors at the phono stage output were originally 2.2µF/25V tantalum types; the replacements were 2.2µF/25V bi-polar types.

Non-signal Path Capacitors:
The filtering capacitors (C84, C85) that shunt the zeners (D12, D13) were originally rated at 220µF/25V and the replacements were 330µF/25V low ESR types. The capacitors (C3) on AWK-065 were originally rated at 100µF/10V; replacements were 100µF/25V low ESR types.

SX-750%20Phono%20Restored%20Label_zpsqybdqydm.jpg
 
Last edited:
It is enlightening to see how the 750 is designed in comparison to the other x50 series. Each model takes on a bit of personality of the designer and the bit of latitude that is allowed by the designer as long as they meet their cost targets.
What stands out to me is the RF section where the designer decided to used a ratio detector as they did in the 1050 and 1250. He could have just as well used the HA1137 quad detector as was done is so many other designs.
I also see that the designer used a boot strapped VAS instead of a CCS type of design.
The phono design change to a discrete design vs the TA7136 is another one.
As always, great work and documentation. I am sure the original designer would be very happy with your work and upgrades.

I would like to ask why you did not use a film for C1,2 on AWG-046? for the sake of reliability more than say performance.

Cheers
 
Thanks for the kind words, Rick. I agree with you regarding the 750, it seems to have benefited from a "trickle down" effect from the bigger x50 models (e.g., 4-gang tuning capacitor like the 850 and 950 rather than the 3-gang one in the 650 and lower).

Although its not obvious from the pictures, the restoration/upgrade of this SX-750 has been a long term project (just wasn't a priority). The first part of the restoration/upgrade, which included the AWR-099, AWG-046, AWM-094 and the phono stage, was completed almost 9 years ago when I first got the unit. This is why these boards do not have stacked film types: predates my use of this capacitor type (not sure they were even readily available then). Currently, the approach is to use stacked film types for capacitors that are ≤4.7µF. Hence the power amplifier board (AWH-046) that was restored last November uses WIMA stacked film capacitors; the 0.22µF stacked films on AWR-099 were installed when the relay was recently replaced.
 
DO you guys use turntables connected to these receivers? Makes any sense to use a preamp or just connect the turntable to the phono input of the receiver?
 
Just wanted to add my thanks to Leestereo for another superbly documented restoration & upgrade thread - I particularly appreciate the detailed discussion of parts choice vs. circuit locations, which is not generally well documented in my (very limited) experience. :thumbsup:

:lurk:
 
DO you guys use turntables connected to these receivers? Makes any sense to use a preamp or just connect the turntable to the phono input of the receiver?

I reccomend using the receiver's phono input. The phono stage in 70s stereo equipment is typically excellent, understandingly so since records were that era's primary medium for high quality sound.
 
I have a few questions for you Lee, (if that is your real name :) why do you prefer to use BP caps, I assume that you are primarily using the Nichicon museES series (green ones) vs Low Leakage KLseries for AC coupling?
I agree for places like AC coupling into circuits with zero or no DC bias voltage, it is probably best to use the BP caps.
iirc mtf states that he uses/prefers KL series because they are lower noise, can you comment if you have noticed this? I guess we could ask mtf if has quantitative poof of his claim :) maybe it comes down to a Coke, Pepsi preference.
I recently bought a QX-9900 to play with and use in my garage loft area. It is a monster to get into some places for service. I am putting together my long list for replacement ecaps, since I have now torn into it. Have you ever worked on a QX-9900? It is very similar to a SX-727.
I want to finalize my order list, determine if I am going to go with ES or KL series ecaps for the signal path parts.
To help me and others, can you specifically state what Wima, Low ESR ecap types that you are using or prefer?

Thanks
Rick
 
In the signal path, my preference is to use film capacitors whenever possible, i.e., for capacitors that are ≤4.7µF. For larger values I will typically install Nichicon ES bi-polar types. This approach is largely derived from the work of Cyril Bateman examining the impact of capacitors on sound quality: https://linearaudio.nl/cyril-batemans-capacitor-sound-articles. He tested electrolytic capacitors extensively and demonstrated that in AC coupling, bipolar electrolytic capacitors have the lowest distortion/noise of all the electrolytic types (even the presence of a polarizing voltage) and that film capacitors are generally lower distortion/noise than electrolytic capacitors of any type. IIRC, @markthefixer preferred low leakage electrolytic types over low ESR types; I don't recall seeing his opinion on the use of bi-polar electrolytic capacitors. I have not personally noticed an obvious increase in noise from the use of low ESR types in the signal path, but have not directly compared a low leakage electrolytic and a low ESR type in the same position.

To date, I have not worked on a QX-9900 (or any of the Pioneer quad models), but am familiar with the x2x series (nice build quality).

As for capacitor preferences, for stacked film I will typically use WIMA MKS2 or Kemet R82. For values <0.1µF: WIMA MKP2, Kemet PHE or C0G types. For low ESR types, there are more choices: Nichicon PW, PM, PS; Panasonic FM, FR; United Chemicon KZE...availability, price and form factor are taken into consideration when deciding what to order.

Ben ("Lee" is my real last name).
 
Hi Ben,

Thanks for the links to Cyril Bateman's page at LA. I am aware of LA, but not these articles. Much to read for me.
It is too bad that he has passed away on us, he was a great contributor.
 
Tuner & AF (AWE-073) Assembly

AWE-073a_zpsymgmb7yk.jpg


Signal Path Capacitors:
The original capacitor (C86) for the output of the TA7061 IC was a 4.7µF/16V bi-polar electrolytic type and it was replaced with the same type (rated at 25V). Similarly, the C45 capacitor at the input of the HA1196 IC was replaced with a 4.7µF/25V Bi-polar electrolytic type; the original was a polarized 4.7µF/16V polarized electrolytic type.

Non-signal Path Capacitors:
The C27 capacitor, which damps the motion of the centre tune meter, was a 0.47µF/50V electrolytic type and was replaced with a 0.47µF/50V low ESR type. Similarly, the C54 capacitor which damps the signal strength meter, originally 220µF/6V, was replaced with a 220µF/25V low ESR type. The local decoupling capacitors (C41, C42 and C62) were originally rated at 100µF/16V, 220µF/16V and 220µF/16V (respectively) and the replacements were 100µF/25V, 220µF/35V and 220µF/35V low ESR types, respectively. The electrolytic capacitors C46, C47, C60, and C61 were replaced with low ESR types at the same capacity as the originals: 1µF/50V, 3.3µF/50V, 4.7µF/50V and 47µF/50V (respectively). The capacitors C80, C81 and C111 were all originally rated at 10µF/16V and their replacements were 10µF/50V low ESR types. The capacitors C48 and C66 were originally electrolytic types rated at 0.22µF/35V and 0.1µF/35v (respectively), these were replaced with film types of the same capacity.

Tuner%201%20restored%20label_zpsi6xsbxo5.jpg



Filters & Muting Assembly (AWM-094)

AWE-073b_zpszrjn6n8p.jpg


Signal Path Capacitors:
The capacitors (C1, C2) that follow the low pass filter (F1) were originally 2.2µF/25V tantalum types and these were replaced with 2.2µF/25V bi-polar types.

Non-signal Path Capacitors:
The local filtering capacitor C5 (originally rated 100µF/35V) was replaced with a 100µF/50V low ESR type (an increase in the voltage rating is recommended since this capacitor is subject to 34V). The capacitors C3 and C7 were originally 10µF/35V and 1µF/50V (respectively); their replacements were 10µF/50V and 1µF/50V low ESR types, respectively.

SX-750%20Tuner2%20Restored%20label_zpsczl5haba.jpg
 
Hi Ben,

A few comments to make on the power amp section.
Similarly, the garden-variety 120pF ceramic capacitors (C5, C6) in the low-pass filters were upgraded to 47pF C0G types (the lower value eliminates any high frequency phase shift within the audible range due to the filter F3).
The -3dB point with R5/6 (2K2) and the C5/6 (120pF) cap, make a 603KHz LPF which acts as an RFI filter, so it has very little phase shift in the audio freq range. Replacing C5/6 with 47pF raises this -3dB point to 1.5MHz, it now is allowing AM band frequencies to get into the amp, if strong enough they can be rectified by the input bjt and potential cause it to act as a AM detector, not a such a good idea imo.

Non-signal Path Capacitors:
The stock local decoupling capacitors, C9 and C10, were rated at 100µF/35V and these were replaced with 220µF/50V low ESR types. The emitter by-pass capacitors (C14, C15) for Q3 and Q4 (respectively) were originally rated at 220µF/6V; the replacements were 220µF/25 low ESR types.
Actually they are in the signal path, as that is part of the VAS (Voltage Amplifier Stage), it is a bootstrapped design. Not to say that using a low ESR is not okay, I just want to clarify for you the way the ckt works.
C14,15 are also in the signal path, since they pass AC around the emitter resistors, block DC, to allow for emitter degeneration bias.
C7,8 are part of the feedback network, so they pass AC as well, make the amp have unity DC gain. AC gain of the amp is set by approx. 1+(R17/R15) or 1+(27/1.1) = 25.5(28.15dB).
This designer has done something that I do not see being done in many Pioneer designs, but is common practice in Sansui designs, this is to bypass a ecap with a film as done with C25,26.

Cheers
Rick
 
The -3dB point with R5/6 (2K2) and the C5/6 (120pF) cap, make a 603KHz LPF which acts as an RFI filter, so it has very little phase shift in the audio freq range.Replacing C5/6 with 47pF raises this -3dB point to 1.5MHz, it now is allowing AM band frequencies to get into the amp, if strong enough they can be rectified by the input bjt and potential cause it to act as a AM detector, not a such a good idea imo.

The increase in the F3 to ~1.5MHz in the LPF is deliberate on my part to reduce the phase shift at 20kHz to <1°. This LPF F3 value is not uncommon as Marantz often specifies a LPF with the same F3 (1kohm/100pF) and I have noted even higher LPF F3 values in other amplifiers that I have restored, e.g., Kenwood M1 (2.3MHz) and Sansui AU-919 (4.1MHz). Also note that the stock Pioneer LPF F3 of 604kHz still theoretically overlaps the AM band lower end (starts at 540kHz). But there doesn't seem to be a consensus LPF F3 value, even in the same brand, e.g., the Sansui AU-717 specifies an LPF with a 723kHz F3, which is coincidentally the LPF F3 of the Pioneer SX-850. And then there are the designs which don't include an input LPF...

Actually they are in the signal path, as that is part of the VAS (Voltage Amplifier Stage), it is a bootstrapped design. Not to say that using a low ESR is not okay, I just want to clarify for you the way the ckt works.

Thanks for that clarification, didn't realize C9/C10 were part of the VAS.

C14,15 are also in the signal path, since they pass AC around the emitter resistors, block DC, to allow for emitter degeneration bias.

This I did notice, but in my posts, I include the emitter resistor bypass capacitors in the "non-signal path" category to simplify the explanations (as per schematics that trace out the direct signal path). Nevertheless, one can consider that even the power supply is in the signal path, since it is essentially in series with the output stage, a concept that I gleaned from a Nelson Pass article: Power Supplies

This designer has done something that I do not see being done in many Pioneer designs, but is common practice in Sansui designs, this is to bypass a ecap with a film as done with C25,26.

Also worth noting is that it is equally common to find a SX-750 that has a power amplifier board without the C25 and C26 film by-pass capacitors (as seen in the AWH-046 board illustration in the service manual).
 
Last edited:
Microphone Input Stage (AWE-073)

SX-750%20mic_zpsxvp6avz5.jpg


Signal Path Capacitors:

The input (C102) and output (C105) coupling capacitors were originally 2.2µF/25V tantalum types and were replaced with 2.2µF/25V bi-polar types.

Non-signal Path Capacitors:

The local decoupling capacitor (C106) was originally rated at 220µF/35V and was replaced with a 220µF/50V low ESR type (an increase in the voltage rating is recommended since this capacitor is subject to 33V).


Power Supply Assembly (AWR-100)

Power%20Supply%20AWR-100_zpssbya7kac.jpg


The capacitor C1 in the power supply for the indicator lamps was originally rated at 470µF/10V and the replacement was a 470µ/25V low ESR type. Note that by re-positioning the wires, there was enough slack to tilt the AWR-100 board out enough to replace the C1 capacitor without undoing the wirewrap connections.

SX-750%20Fuse%20Board%20Restored%20label_zpsbwrbqmuf.jpg


The original 8V/300mA dial illumination lamps (PL1, PL2 and PL3) on the 7.5V AC secondary were replaced with 8V/70mA LED bulbs (note that due to the multi-LED construction, there is no light flicker even though they are on an AC circuit).

Triac%20amp%20LED%20label_zpsfbjjme3x.jpg


Power Switch/Triac Modification

The rotary power switch in the SX-750 is a known weak spot and although the one in this particular unit was still in very good condition, the @K7sparky triac power switch modification was installed as a preventative measure.
 
Last edited:
Back
Top Bottom