Brilliant! thanks for share!
BA312 replacement
- Thread starter bubhardy
- Start date
That is great news!
Could you hear any difference in noise floor?
Could you hear any difference in noise floor?
Nice work, aldena007!
I got my selector and tape copy switches, and volume/balance pot cleaned up nicely from my 9090. Getting ready to reinstall. I took the opportunity to deoxit the switches and faderlube the pots on the front panel boards, with all the guts hanging out, and such, before I test.
I’ll yank the bad BA312 out next, and sub in my trial replacement. Same circuit, so I expect the same results that you got. It’s nice to know that the circuit works well. I’m probably going to do my own layout. I have this idea to bend the transistor leads 90 degrees, and mount the transistors along the top edge of the board.
On the noise floor question. The test 9090 was one that I have been working on for several months. One thing I had been able to do was to get the unit dead quiet while with no audio signal in play. The BA312's did not create any detectable noise to my ear. I would expect this since the BA312 clone is just a reverse engineered circuit of an old design. There is nothing new in the clones that would impact the noise floor that the original BA312 IC design can or cannot do. The SMD version seemed to improve the phono input a little but I only evaluated the clone by my own hearing. The playback of the LP was excellent. I think the best the clone can do is duplicate the original IC performance which it easily does.
Just curious-when you listened to the unit for noise from the phono stage, did you short the phono inputs to ground?
I expect the performance would be fairly equal, maybe a tad bit better, using the KSC1845.
Hopefully, later today, I will get a chance to power up my 9090 with my little “Frankenstein” BA312, and see how it works. I got the switches and pot reinstalled, as well as the input board. First, I have to test for equal signal on both channels up to the BA312 stage, then see what those outputs look like. I still have the original BA312 in one channel, and my thing in the other.
Here’s a pic of where I’m at-
The towels are so that I can power up disassembled, and still have access to the boards that I need to test.
Here’s the “thing”, installed-
Gotta take my wife shopping now,
My 9090 is not dead quiet. What did you do to get there?
I found that the schematic indicates the grounds without any mention that it is important to ground all the preamp grounds are to chassis ground. The grounding on the power supply grounding pads will have noise on them from the filter capacitors and the rectifiers. The AC power for the lamps is also noisy [60 cycle hum/noise]. Check the grounding pads on the power supply board and you can detect it. I use a small battery powered amplifier [a Radio Shack item]. The process is to have a probe to the test amp input and the test amp ground to chassis. Probe the ground pads with test amp and you will easily hear the noise on the power supply board. You will get plenty of noise on the filter cap ground. My experience is this is the source of much of the 60 Hz artifacts. The tone control circuit is comprised of three or four boards with the ground is "daisy chained" together then a single ground that needs to be grounded to the chassis. If that ground is connected to the same pad that the filter capacitors use you will get noise feedback into the tone control system. Sansui mentions in a service bulletin to make a ground lug to the chassis for the preamp boards.
The small amplifier was available from Radio Shack but no longer available. You can find them on eBay occasionally. The probe uses a capacitor that ties the input to the amp and ground [0.022 mfd 200 Vdc] which I think is a Wyma brand. You can also modify a oscilloscope probe to do the same thing. This setup makes the noise audible and is simpler than a oscilloscope to set up. If you have 60 Hz feedback this can help track down the source. The first thing I do is to power up the DUT and then find the pre-amp/tone board ground and disconnect it from the existing ground. If the audio artifact is coming from the power supply ground pads it will go away when you disconnect the ground wire. You then can try alternate grounding to the chassis and see if your artifact goes away. If it does you are good. Audio artifacts can come from other sources such as noisy components but this rig will help you find them. If you have a dead channel you can inject a signal and trace it until it dies. I found a bad BA312, Dolby IC, and bad/noisy transistors with this rig. I will look around and see if I can find a schematic on how to build the probe. The Radio Shack amplifier has a volume control so you can track down faint noise on the pre-amp side which gets amplified by the power amp side. This is my go to rig when it come to audio artifacts. I have "cleaned up" audio artifacts on several 9090's and 9090DB's with this device.
The small amplifier was available from Radio Shack but no longer available. You can find them on eBay occasionally. The probe uses a capacitor that ties the input to the amp and ground [0.022 mfd 200 Vdc] which I think is a Wyma brand. You can also modify a oscilloscope probe to do the same thing. This setup makes the noise audible and is simpler than a oscilloscope to set up. If you have 60 Hz feedback this can help track down the source. The first thing I do is to power up the DUT and then find the pre-amp/tone board ground and disconnect it from the existing ground. If the audio artifact is coming from the power supply ground pads it will go away when you disconnect the ground wire. You then can try alternate grounding to the chassis and see if your artifact goes away. If it does you are good. Audio artifacts can come from other sources such as noisy components but this rig will help you find them. If you have a dead channel you can inject a signal and trace it until it dies. I found a bad BA312, Dolby IC, and bad/noisy transistors with this rig. I will look around and see if I can find a schematic on how to build the probe. The Radio Shack amplifier has a volume control so you can track down faint noise on the pre-amp side which gets amplified by the power amp side. This is my go to rig when it come to audio artifacts. I have "cleaned up" audio artifacts on several 9090's and 9090DB's with this device.
My BA312 temporary replacement works, too.
Cleaning the selector switch and the tape copy switch, as well as the volume/balance pot, worked wonders. I have a nice clean signal going into the BA312s on the tone board, all the way through to the driver board inputs.
Next, I will be rebuilding all of the front panel boards-F2542 through F-2545, while I have them out. The driver board, the protector board, and the power supply board were already rebuilt, so that will leave the F-2541 equalizer (input) board to rebuild.
Had to fix the typos.
Cleaning the selector switch and the tape copy switch, as well as the volume/balance pot, worked wonders. I have a nice clean signal going into the BA312s on the tone board, all the way through to the driver board inputs.
Next, I will be rebuilding all of the front panel boards-F2542 through F-2545, while I have them out. The driver board, the protector board, and the power supply board were already rebuilt, so that will leave the F-2541 equalizer (input) board to rebuild.
Had to fix the typos.
Last edited:
Fired up my 9090 tonight. Front panel boards back in-all rebuilt. Fixed a couple of wires that broke off the front panel boards, in the course of work. Installed the new filter caps. Nichicon LKG, 10000uF@80V. 63VDC at idle. Nice! Going to run the SG into the aux input, see how things look on the scope.
Last edited:
Sure did. They look the same at 1kHz, same gain. Signal is clean, in to out. The boards fabbed from my layout, from JLBPCB, should be here today.
I found that I had the rotor 180 out on one wafer of the input selector switch to fix. So I decided to recap the F-2541 input board, too, while I was in there.
But there’s something else screwy going on. I get much more output level on one channel than the other. Got to find out why! This 9090 was a real pigsty to begin with, and in protection. So I’m not surprised. It had many issues to deal with-and I haven’t checked out the tuner yet!
I found that I had the rotor 180 out on one wafer of the input selector switch to fix. So I decided to recap the F-2541 input board, too, while I was in there.
But there’s something else screwy going on. I get much more output level on one channel than the other. Got to find out why! This 9090 was a real pigsty to begin with, and in protection. So I’m not surprised. It had many issues to deal with-and I haven’t checked out the tuner yet!
This 9090 seems to work fine now. Measures 136WRMS/139WRMS output. Only thing is, there’s no FM. With a random wire antenna, there’s nothing. Meters don’t move. I know that the signal strength meter works on AM. FM section has DC power going to it-just no output.
I got the BA312 discrete boards in from JLBPCB. Look good. I need to build some up, stick a couple in, and scope it.
I got the BA312 discrete boards in from JLBPCB. Look good. I need to build some up, stick a couple in, and scope it.
Planning to sell any of those once completed?
I still have to build up some and test them.
The schematic that I used is same one that Tupolevfan posted, and the one that I hand wired worked great.
I wasn’t planning on selling them.
But, if you PM me your address, I can drop some in the mail to you, @Magnus Back, along with a BOM and schematic.
The schematic that I used is same one that Tupolevfan posted, and the one that I hand wired worked great.
I wasn’t planning on selling them.
But, if you PM me your address, I can drop some in the mail to you, @Magnus Back, along with a BOM and schematic.
Last edited:
I just saw this item on the big auction site. It looks interesting and at $20.00 the price is right. I ordered two of them and when they arrive next week I will install them in a Marantz 2275 I am restoring that has a bad BA312. I will post the results once I receive and install them. Is anyone familiar with this particular device?
Thanks!
Thanks!
I used a few boards @bubhardy sent me and they work great!
(See above).
This is the equivalent circuit:
Where in Michigan are you, @RichieM?
A quick update on the 9090 above. AM works, but FM is stone silent. The PS voltages shown in the schematic are good. The front end has PS. I even tried replacing the FET in the front end, but nothing. It is now in the magic hands of @Punker X. He has made the tuners in an 8080, 4000, 5000, and 5000A sound the way that I recall they did, when this gear was new, and I was MUCH younger. 
I hope he can get the 9090 FM working.
I hope he can get the 9090 FM working.