I drilled & tapped an M3 hole in the plate. Its easy enough to do, but use some lubricant when tapping the hole, and take care. The bigger challenge for me was cutting the box down to size. It started off as the 230mm long variant I used to have my DAC PSU in. Amazing what you can do with a hacksaw and years of practice 😉. Its great to recycle...
Troubleshooting has started. Cleaned up questionable soldering, found C12 in backwards. The bulging top was a clue... Left channel works, right does not. U17 has -15v on pin 2, U7 has 0V. Which is correct? I don't see a DC path to the previous stage, and voltages on U27\U18 match. Another issue noticed in testing is with the MC auto 0 jumpers open the voltage at the MC input varies between +175 and -175mv, with less change every cycle until settling near 0. With the jumpers in place it is +30mv to -30mv, settling near 0 after ~10 seconds. Should I have a MC cartridge plugged in, or a load across the inputs while testing? I once was very good at troubleshooting, 25+ years ago. A bit rusty now.
Bob, here are some answers for your questions.
MC Input.
The voltage measured at the MC input wrt 0V will start at some level other than 0 (typically up to +- 30mV) and then settle to 0V as the servo loop takes control. This is normal.
Sorry, I don't understand the +-175mV part of your question - that is too high whether the servo is connected (U28 and U29 flashed short) or not (U28 & U29 OPEN). Typically you will read up to +-30mV on the input with the flash pads open. After you connect them, measurements should be +-25uV max, +-6uV typical.
For the MC input offset setup and test, the inputs must be left OPEN.
Note that you must measure these voltages directly across the input connector. If you measure between the central 0V and the signal input, you will get an error due to the mV drops across the PCB traces etc. The servo takes its 0V reference from the MC input connector signal ground.
Rumble/Acoustic Filter
The voltage on pin 2 on both U7 and U17 must be 0V under normal operating conditions. I would check around pin 2 on U17 - looks like you have a short to -15V or U17 is damaged. The output voltages on U18 and U27 (with the MM input shorted) must also read 0V +-3mV.
For the MM input, I would short these out while you do the tests to prevent noise pickup which might give you erroneous readings on your meter.
Please keep us updated on your progress.
🙂
MC Input.
The voltage measured at the MC input wrt 0V will start at some level other than 0 (typically up to +- 30mV) and then settle to 0V as the servo loop takes control. This is normal.
Sorry, I don't understand the +-175mV part of your question - that is too high whether the servo is connected (U28 and U29 flashed short) or not (U28 & U29 OPEN). Typically you will read up to +-30mV on the input with the flash pads open. After you connect them, measurements should be +-25uV max, +-6uV typical.
For the MC input offset setup and test, the inputs must be left OPEN.
Note that you must measure these voltages directly across the input connector. If you measure between the central 0V and the signal input, you will get an error due to the mV drops across the PCB traces etc. The servo takes its 0V reference from the MC input connector signal ground.
Rumble/Acoustic Filter
The voltage on pin 2 on both U7 and U17 must be 0V under normal operating conditions. I would check around pin 2 on U17 - looks like you have a short to -15V or U17 is damaged. The output voltages on U18 and U27 (with the MM input shorted) must also read 0V +-3mV.
For the MM input, I would short these out while you do the tests to prevent noise pickup which might give you erroneous readings on your meter.
Please keep us updated on your progress.
🙂
So... I hooked up a signal generator and a scope. Found the LSK389 L Ch 22 ohm G-D, Right channel 0 ohm G-D both sides. Maybe a mistake converting my TO-71 lsk389 pinout to SOIC? I have ordered SOIC LSK and will try that when they get here... Also ordered OPA1641, there are no LM4562 in the US currently. Shotgun is my middle name....
Bob, The PCB can take SOIC or the TO-71 but make sure you don’t have any shorts on the PCB tracks between the pads - very easy to do.
Unfortunately there is a world wide semiconductor shortage at the moment. I thinks it’s the worst allocation in the last 20 yrs.
(Fine tip soldering iron and 1 mm or less solder wire a must!)
Unfortunately there is a world wide semiconductor shortage at the moment. I thinks it’s the worst allocation in the last 20 yrs.
(Fine tip soldering iron and 1 mm or less solder wire a must!)
LSK389B or LSK389C
Question on the LSK389B or LSK389C
Hope that somebody can give me some information about the use of the LSK389. Can I use the LSK389C instead of the LSK389B in the X-Altra MC/MM preamplifier design.
Question on the LSK389B or LSK389C
Hope that somebody can give me some information about the use of the LSK389. Can I use the LSK389C instead of the LSK389B in the X-Altra MC/MM preamplifier design.
So I replaced the LS358 with SOIC and now I have output on both channels. It is equal up to the final output IC, but the output to the jacks is ~30% lower on the right channel. Can you clarify that the Turnover switch should short 2 to 1 and 3 when switched? I have something wrong in the final stage....
Bob, please see this diagram which shows how the turnover switch is wired
http://hifisonix.com/wordpress/wp-content/uploads/2021/04/Filter-switch-wiring.pdf
The switch adds in or removes an extra bank of capacitors to shift the turnover frequency between 20 Hz and 45 Hz.
http://hifisonix.com/wordpress/wp-content/uploads/2021/04/Filter-switch-wiring.pdf
The switch adds in or removes an extra bank of capacitors to shift the turnover frequency between 20 Hz and 45 Hz.
I saw that, and used it to wire the switch. It does not show the electrical diagram though. In my testing, 2 shorts to 1 and 3 when the switch is up. Is that correct? On the weak channel the waveform at point 1 distorts somewhat when switched on with a 1khz input. The other channel does not. I will measure the components in that area, and maybe replace the final OPA1641 in the weak channel. Almost there....
You are correct. 2 simultaneously connects to 1 and 3 when in the up position. In the down position, 1 and 3 are open.
Last edited:
I wanted to report back that the final problem was U17, the one that had -15v on pin 2. One channel output is ~5% lower than the other but I can live with that for now....
Hello Bob, let me come back to you later today. I am out and about at the minute. Both channels must give exactly the same output.
With the MM inputs shorted and fed an arbitrary level 1khz signal the difference is first visible at pin 7 of the LM4562 ICs. Left channel .58 vpp, right channel .53vpp. At the outputs 1.7vpp and 1.55vpp. Maybe mismatched LSK389s? I would swap them but the PC board may not survive another bout of surgery. A visual check shows all resistors are the correct value. Caps were installed one value at a time so probably are correct.
I think it’s premature to assume it’s the LSK389’s at this stage 🙂
Can you inject a c. 5 mV square wave and then tell me if the output square waves on pin 7 look EXACTLY the same - perhaps put a picture up.
Can you inject a c. 5 mV square wave and then tell me if the output square waves on pin 7 look EXACTLY the same - perhaps put a picture up.
Bob, not you will not get a square wave out due to the EQ ( I should not have said ‘look at the square wave on the output - my bad!) but more like a triangular shape. But what we are looking for is to see if they are the same. If they are not, then the problem will lie in the EQ network probably.
Also check the voltage drop across the drain load resistors and the source resistors.
Also check the voltage drop across the drain load resistors and the source resistors.
