Leestereo's Restoration and Upgrade Of A Marantz 2270

[Leestereo]

A nice 2270 was on the bench for restoration/upgrade of the power supply, preamplifier, power amplifier and phono stages.

Main Power Capacitors
The original Elna 9,000µF/55V capacitors were replaced with Kemet 15,000µF/63V capacitors. Note that the original capacitors measured ~11,800µF at time of removal (vintage capacitors were often specified as -10%, +50%).

The new capacitors were installed with a "star" ground (to minimize noise), instead of having the individual ground wires attached to the buss bar between the main capacitors in the original configuration. Also, the rf capacitors on the bridge rectifier were replaced with same value (0.22µF) XY safety capacitors.

 

[Leestereo]

Power Supply & Relay Board (P800)
The P800 board contains the regulated power supplies as well as the protection circuit.

The S1B-01-02 diodes (H801, H802 and H811-H813) were upgraded to UF4005 soft recovery types. The C801 capacitor is the main filter for the regulated supplies and the original 470µF/63V capacitor was replaced with a 1000µF/63V long life, low ESR type. Similarly, the 330µF/50V capacitors C802 and C804 which filter the 13.5V and 35.7V supplies, respectively, were replaced with 680µF/63V low ESR types. Capacitor C803 (470µF/16V) which shunts the 14V zener, was replaced with a 470µF/35V low ESR type. The 3.3µF/35V capacitor (C805) on the base of the H806 regulator was replaced with a 10µF/50V low ESR type. Capacitors C806, C807 and C810 are used in the protection relay circuit and were also replaced with low ESR types: 22µF/50V, 220µ/35V and 47µF/25V, respectively. The protection relay (L801) was replaced with an OMRON MY2-DC24, which is a drop-in replacement for the original type.

The large 150 ohm/5W dropping resistor, R801 (installed on chassis), that feeds C802, was replaced with a physically smaller 150 ohm/7W vitreous enameled type (with low TC).

 

[Bert 1100]

 

[Leestereo]

Preamplifier (P400) and Tone (P450) Boards

The 0.22µF polyester film capacitors in the high-pass input filter of the preamplifier, C401 and C402, were upgraded to 0.33µF WIMA MKP2 polypropylene films; the larger capacitance value lowers the F3 to ~5Hz (minimizes phase shift in midrange of the passband). The polarized electrolytic capacitors in the signal path at C403/C404 and C407/C408 and C410/C411 and C414/C415 were repalced with polyester stacked film types of the same capacity as the originals.

The local DC filtering capacitor for the P400 board (C409), was originally rated at 100µF/35V and the replacement was a 270µF/50V low ESR type. Similarly, the filtering capacitors, C405 and C406, that were originally 10µF/35V were replaced with 47µF/50V low ESR types.

In the 2270, it is worthwhile to upgrade the capacitors on the tone control board since the tone controls cannot be "defeated" and hence are always in the signal path (regular ceramic capacitors are particularly poor sounding). The polyester capacitors at C451-C456 and C457/C458 were upgraded to polypropylene types of the same capacity as the originals. The signal path ceramic capacitors at C457/C458, and C461/C462 were upgraded to C0G types of the same capacity as the originals. Note that in this particular unit, the C459/C460 ceramic capacitors were NPO (C0G) types and hence did not need to be replaced.

 

[Hipocrates]

 

[Steven Tate]

+1. This is going to be one of those gold standard threads. I have a 2270 in the wings. I will be watching with great interest.

 

[restorer-john]

IMO, the earthing strategy you have decided upon is not remotely ideal.

You have created a considerably higher impedance for all the ground/return currents (a single jumper wire to the cap common is now carrying all the current) and you are likely going to introduce eddy currents in the chassis than were not present before.

The original method employed offered the lowest potential for this to happen, with all high current returns tied to the cap common 0v and the chassis not being subject to large return currents.

There are many instances where the deliberate use of a controlled (known) impedance earth strap to chassis were implemented specifically to remove stray chassis currents and noise. Denon, for example used this method extensively in their TOTL power amplifiers.

 

[Leestereo]

IMO, the earthing strategy you have decided upon is not remotely ideal.

You have created a considerably higher impedance for all the ground/return currents (a single jumper wire to the cap common is now carrying all the current) and you are likely going to introduce eddy currents in the chassis than were not present before.

The original method employed offered the lowest potential for this to happen, with all high current returns tied to the cap common 0v and the chassis not being subject to large return currents.

There are many instances where the deliberate use of a controlled (known) impedance earth strap to chassis were implemented specifically to remove stray chassis currents and noise. Denon, for example used this method extensively in their TOTL power amplifiers.

I am not sure I fully understand your argument (possibly because I am not an EE, just a hobbyist). The "star" ground approach adopted for the 2270 described herein is derived from Audio Power Amplifier Design Handbook (5th Edition) by Douglas Self (page 196): "...the star point A is defined on a short spur from the heavy connection joining the reservoirs; trying to use B as the star point will introduce ripple due to the large reservoir charging current pulses passing through it."

Figure 7.4: Distortion 5: Correct way to route decouple grounding to the star-point.

 

[restorer-john]

The star ground is perfectly valid, but the issue as I see it here is, you will end up with a voltage drop caused by high current flow between the tied cap terminals and the chassis via the short link of wire, as you have moved the centre tap (the highest current carrying cable in the whole amp) to the chassis point. The ripple you are hoping to avoid is now all flowing through your wire link.

If you look at the Self's drawing, he doesn't propose moving the centre tap of the transformer to the star point- it is still connected as per normal.



If you move the transformer centre tap back to the capacitor common bus bar and keep the rest of your arrangement, you may reap the benefits of his star arrangement.

It all depends on what other 'cheeky' chassis earths there are floating (pun intended) around in the 2270.

 

[Leestereo]

The star ground is perfectly valid, but the issue as I see it here is, you will end up with a voltage drop caused by high current flow between the tied cap terminals and the chassis via the short link of wire, as you have moved the centre tap (the highest current carrying cable in the whole amp) to the chassis point.

If you look at the Self's drawing, he doesn't propose moving the centre tap of the transformer to the star point- it is still connected as per normal.

View attachment 1034244

If you move the transformer centre tap back to the capacitor common bus bar and keep the rest of your arrangement, you may the benefits he espoused.

OK, misinterpreted that part of the D. Self diagram; will look into moving the transformer center tap to the buss bar. As always, the devil is in the detail!

 

[sherod]

I increased the capacitance on the P400 board for the filtering caps, C409, from 100uf to 220uf and C405,406 from 10uf to 33uf, all just slightly lower than what Lee increased them too. I am hearing a very nice improvement in soundstage, better dynamics and deeper and more precise bass. Thanks for the suggestion, Lee.

 

[Leestereo]

Phono Board (P700)

The stock capacitors in the input high-pass filter (C701, C702) were 0.47µF polyester film capacitors and these were upgraded with 1.0µF polypropylene film types. The polarized electrolytic capacitors C703/C704, which determine the low end roll-off of the RIAA equalization, were originally 22µF/6.3V tantalum types. Their replacements were bipolar 47µF/35V Nichicon ES capacitors; the larger capacitance lowers the F3 to ~5Hz (minimizes phase shift in midrange of the passband). The stock local filtering/decoupling capacitor, C713, was a 100µF/50V was replaced with a 270µF/50V low ESR type. The original polyester film capacitors at the output, C714/C715, were upgraded to 1µF polypropylene film types. The failure-prone 2SC458 transistors (H705, H706) were replaced with KSC1845 transistors. The 1S-2473 epoxy diodes (H707, H708) were replaced with 1N4148 diodes and the failure-prone VD1212 dual diode (H709) was replaced with a pair of 1N4148 diodes in series.

 

[restorer-john]

Looking good!

270uF is a funny value huh? you almost never see it in vintage. It's usually 100/220/330/470/1000uF. 680uF seems to have disappeared too, not sure why, there's a big hole between 470 and 1000. 6.8/68/680 was dropped by many makers a while back.

 

[Leestereo]

Power Amplifier (P750)

The stock capacitor in the input high-pass filter (C751) was a 1.0µF polyester film type and it was upgraded to a polypropylene film type. The C753 (1000pF) and C754 (500pF) capacitors were ceramic types that were upgraded to C0G types (1000pF and 470pF, respectively). The 15V zener diodes H762 and H763 were upgraded to low noise 2% variants and the capacitors C759 and C760 that shunt the zeners, were originally rated at 10µF/16V and were replaced with 47µF low ESR types. The C761 capacitor, which determines the negative feedback low end cut-off, was originally a tantalum type and was replaced with a bipolar 33µF/25V Nichicon ES capacitor. The feedback resistors R757 and R759 were upgraded to 1% metal film types. Note that ceramic capacitors in the signal path that were NPO (C0G) types originally need not be replaced.

 

[Steven Tate]

Question about the 15V zeners. Would a 1/2 watt version be adequate for these?
Thanks!
Steve

 

[Leestereo]

@Steven Tate: 0.5W zener diodes are not recommended in this application; the stock 15V zeners are rated at 1W and the replacements (BZ85B15) are rated at 1.3W.

 

[Steven Tate]

@Steven Tate: 0.5W zener diodes are not recommended in this application; the stock 15V zeners are rated at 1W and the replacements (BZ85B15) are rated at 1.3W.

Thanks! I saw those 1.3W zeners when I did the Mouser search. Just wanted to be sure which ones to get as I will be doing another 2270 soon. Thanks for this thread. It's very timely for me. I have done another 2270, but I was doing a lot of stumbling around in the dark. AKers bailed me out a few times. This thread will hopefully pave the way for a smooth rebuild.

 

[Leestereo]

Flyback Diodes on P750

A pair of diodes (UF4004) were installed on the back of each power amplifier board (P750) to function as flyback diodes for the output stage.

 

[MBuras]

Nice work as always. Looking very similar to the last 2270 I did for a friend including the flyback diodes.

 

[sherod]

Any opinion on why the Marantz engineers didn't put flyback diodes for the output stage?