PA1002A workaround / replacement

[Bob@FM]

I'm looking at making something work for the PA1002, and have some questions.
First, does anyone know what sort of solid state relay connection is used internally? Does it open the signal path, or short it to ground, or do both? They refer to grounding the signal in the service manual circuit description, but since it has DC on both IN and OUT terminals, I think they must also switch the signal path open. Anyone know?

I'm leaning toward only grounding the audio signal to mute, which requires a little different strategy. I'm not a fan a sending signals through CMOS switches if it can be avoided.

Point is, I prefer to try and make the replacement sound better than stock, not the same. Also it should allow op-amp swaps. Looks like standard single or dual op-amps will work fine here. For the mute switches, I'm looking at Form A SS switches with lowest ON R. They need resistance in front to work as a mute though, so will likely be moved to another area.

Any thoughts or comments from people that have already tried this are welcome.

 

[rcs16]

I do not think there is anything wrong with a old Sanyo LC4966 cmos switch especially if it is feed thru a resistor to a opamp input in a inverting config. The crown FM1,2 have a ckt for that, but they use the lower performing 4066
another option is the way Yamaha does it in the T-85 with those hign Vbe reverse break down bjts.
I am not sure why mute ckts require a huge amount of attenuation, just enough is required to mute the interstation noise so it does not blast your ears out.
There is one thread here where a member made his own version. need to search, but i am going out now.
I can help make a small pcb if you want or need.

 

[Bob@FM]

Thanks for the pointers. I don't see anyone that has made a replacement and posted any details about how it was done, or what parts were used. No biggie, I'll design it, test it, and see what happens.

My main concern is getting the signal out with proper de-emphasis and separation. Muting is not a big deal for me, I use many tuners without it. But It would be nice if not too difficult. I am pretty sure I'm going to do this as two separate modules (hand made to start). One for the amplifier, one for the muting. Mostly as the muting need to be in a different place the way I'm doing it.

BTW, the toughest thing about a muting design is getting it to work without pops. Agreed, absolute silence is not needed, but most work that way, just because... You can get to -50 or -60 dB without a lot of trouble, so that's what I'll do to start. I'll shunt the signal to ground only for muting.

This mute switch can't directly follow an op-amp output, you need a current limiting resistor in front of it, and also a DC block cap for good measure, to avoid pops. That means it need to be physically over in another spot on the board, not where the PA1002A pins 4 to 5 place it. No big deal, the Pioneer construction allows it to be placed on board edge, working to ground signals on the board edge pins. Easy that way.

The engineering is picking the right parts, and getting the interface to the mute signal (levels and current) working OK.
BTW, the other thing to work out on the amplifier, besides pin placement, is the + input bias voltage, usually 1/2 the rail supply voltage. Since Pioneer used active devices for that, I will too, but it would likely work with 2 resistors as well (with compromises in power supply rejection).
Bob

 

[rcs16]

Good luck Bob, I'd like to follow along on your progress.

 

[Bob@FM]

Good luck Bob, I'd like to follow along on your progress.

I'm not going to do pretty and nice drawings, but if you want to follow my sketched ideas, I can post them here.

Also, would like to hear from people just how important muting is for them. If it doubles the price, would some rather have a lower cost board that makes the tuner play again, but without muting?

 

[Bob@FM]

Attached is a drawing showing the elevation profile of my solution. You can do this easily on a perfboard and wire it pin to pin (like a tube audio gear). Or build a board and make a bunch. But let's prove it works first.

 

[Bob@FM]

Here's my sketch for pin to pin wiring of the PA1002 adapter board
It uses a common dual op-amp like an OPA2134. But socketed so I can swap others in and listen for what works best.

I think I already mentioned this design is open, and anyone can use it or copy it. Just please give me credit if you do.

 

[Punker X]

Applaud the effort. Glad someone is finally tackling this. I'll be in line to buy some of the finished boards to have on hand for repairs and upgrades. Once had a customer with a bad one in a SX-1250 and used the part out of a SX-1050 to get the 1250 up and working.

 

[1MEGΩ]

Applaud the effort. Glad someone is finally tackling this. I'll be in line to buy some of the finished boards to have on hand for repairs and upgrades. Once had a customer with a bad one in a SX-1250 and used the part out of a SX-1050 to get the 1250 up and working.

You must be confusing the 50 series with the 80 series units. The SX-1050 and SX-1250 do not use this chip.

 

[Punker X]

You must be confusing the 50 series with the 80 series units. The SX-1050 and SX-1250 do not use this chip.

True that.. was a long time ago... been a while since I've had either a 50 or 80 series across the bench.

 

[tjg79]

I've got an SX-980 that uses that Pioneer IC and I've seen it online that the IC KB4438 is a direct replacement for the PA1002-A. This old thread seems to confirm the cross, but I don't see where anyone has confirmed a successful substitute.

http://audiokarma.org/forums/index.php?threads/pa1002a-module.599206/

Sierra IC Inc appears to indicate that they've got 58 IC KB4438 in stock, but they are sitting in Europe with a $750.00 minimum and they only sell business to business.
https://www.sierraic.com/KB4438

I'm interested to know if anyone has successfully replaced a PA1002-A with the KB4438. If it is a direct substitute, it would appear to be an easier solution than fabricating a replacement from discreet components.

 

[rcs16]

Bob@FM has a thread on this subject going on at fmtunerinfo.

 

[Bob@FM]

Yes, let's spread the fun around. I have a KB4438 on order from little diode (ePay) to see if they are real, and match the original PA1002A for specs and sound quality. While on the subject, the PA1002A is also apparently also identical to the PA4004.

Anyone else have an inputs or comments on the need, or not, for muting?

The PA1002A basically does 2 things - it's a gain stage/buffer, and does the FM tuner muting. Doing a replacement for the gain stage is fairly easy, including the muting is more complex.

BTW, for me, 90% of the reason to do this is to have more amplifier options for the gain stage. Which are hopefully better sounding than the one in the 70's era PA1002A. If you design the replacement board with a standard dual op-amp (w/socket), you get to play with different devices and listen to the difference. That's a fun game in itself, and the TX-9500II deserves the best - be it Burson if you will.
https://www.bursonaudio.com/products/supreme-sound-opamp-v6/

 

[markthefixer]

I've got an SX-980 that uses that Pioneer IC and I've seen it online that the IC KB4438 is a direct replacement for the PA1002-A. This old thread seems to confirm the cross, but I don't see where anyone has confirmed a successful substitute.

http://audiokarma.org/forums/index.php?threads/pa1002a-module.599206/

Sierra IC Inc appears to indicate that they've got 58 IC KB4438 in stock, but they are sitting in Europe with a $750.00 minimum and they only sell business to business.
https://www.sierraic.com/KB4438

I'm interested to know if anyone has successfully replaced a PA1002-A with the KB4438. If it is a direct substitute, it would appear to be an easier solution than fabricating a replacement from discreet components.

I can confirm that I have seen many SX-1980 tuners with KB4438 (and in other chip function substitutes, the KB4437(?) ) chips operating normally - APPARENTLY FROM THE FACTORY!!!! (from the condition of the solder etc on the solder side of the board)
I have done enough of them to be completely certain of all of this.

 

[tjg79]

Thanks for that information mark. I suspect the Pioneer PA1002-A and the KB4438 are the same chip other than labeling; one brand name and the other generic. I see the KB4438 in the NEC AUT-8300E stereo tuner, which I think was manufactured during the same time period as the Pioneer SX receivers that employ the PA1002-A. There may be other still available generic IC substitutes for the other unobtainium Pioneer ICs as well. The key is to look at the service manuals for the electronics that employ the KB4438, and then compare the other ICs and their circuits.

 

[Bassblaster]

Here's my sketch for pin to pin wiring of the PA1002 adapter board
It uses a common dual op-amp like an OPA2134. But socketed so I can swap others in and listen for what works best.

I think I already mentioned this design is open, and anyone can use it or copy it. Just please give me credit if you do.

Are you working on a PCB by chance? i have a bit of free time and i could make a PCB layout.

 

[Bob@FM]

Are you working on a PCB by chance? i have a bit of free time and i could make a PCB layout.

After shipping a few hundred LM4500 boards all over the world, I'm not interested in personally being involved in making, selling/shipping any more new board designs. But more than happy to design, build protos, and test them.

So yes, I'd be happy to work with you to make a layout, and get them made somewhere. I just hand wired up a little adapter for testing. I can also make PCBs here from scratch - layout, laser print, etch, drill, etc. But for something like this, easier and faster to build hand wired. I put a pic up "IF" it works later. I use Eagle for the layout. Have made maybe 30 home etched boards here by now. It gets messy in winter with the acid and all.

 

[Bob@FM]

Tried my prototype of this idea for PA1002A "amp" replacement last night.

It worked great, and sounded outstanding. This restored my TX-9500II to working status again, albeit with no muting.
Honesty, I could live with it forever this way. This sounds more like what I heard from the TX-9500II driving an outboard SS MPX adapter. Better, that is.
So this design works fine as a high end, better sounding PA1002A mod/replacement where one can swap op-amps and use whatever you like.

But I will pursue muting as well, since I already 90% designed it, and bought all the parts needed to do it. One step at a time. As before, I intend to not pass any signals through solid state switches for muting. It will be signal to ground via optical coupled cmos switches that look like ~5 giga-ohm when not open.

 

[Bassblaster]

After shipping a few hundred LM4500 boards all over the world, I'm not interested in personally being involved in making, selling/shipping any more new board designs. But more than happy to design, build protos, and test them.

So yes, I'd be happy to work with you to make a layout, and get them made somewhere. I just hand wired up a little adapter for testing. I can also make PCBs here from scratch - layout, laser print, etch, drill, etc. But for something like this, easier and faster to build hand wired. I put a pic up "IF" it works later. I use Eagle for the layout. Have made maybe 30 home etched boards here by now. It gets messy in winter with the acid and all.

I need a clearer schematic. and some other tidbits. What value is the cap on pin 16 and what about those 2 transistors. pin outs?

 

[Bob@FM]

I need a clearer schematic. and some other tidbits. What value is the cap on pin 16 and what about those 2 transistors. pin outs?

OK, will draw up something and post it later.
The board will need to match the existing PA1002A 16 pins on the bottom, and accommodate the op-amp and rail splitters on top. The 3 pin TLE2426CLP rail splitters do look like TO-92 transistors, but are not. But do use the TO-92 common holes in a the PCB software's component library. And they will need power and ground connections, I'll show all this.

The board will need to be compact to fit into the existing space. Also we will need to select the pins that go from the board to the existing PA1002A holes, machined pins if you want to fit them into a machined pin socket there (recommended). I looked at Mouser, Milli-Max sells many like this.
On the op-amp, only 3 of the 8 pins are an exact match to the PA1002A, so the op-amp pins will need to be offset from the boards pins to the PA1002A.

Also will need a terminal to supply the mute signal from PA1002A pin 8 to another small muting board. This will make more sense once you see the better drawing.

I like to stay with big through hole parts, for ease of construction. But it could also be done all in surface mount, making it even more compact, and maybe even fit on muting as well. But only if someone else builds them who is setup for that. SM is usually not for most hobby guys that occasionally solder.