• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Yet another butchered 250

ocdecio said:
It's cheap and available:

https://www.mouser.com/ProductDetail/central-semiconductor/mpsa20-pbfree/?qs=u16ybLDytRbwdBgWKS1Zxg==&countrycode=US&currencycode=USD

I wouldn't be too concerned with fT in this application. Power should not be a big issue either.
Click to expand...
Yes, I saw those at Mouser, and I already have an order ready to go in so will add some MPSA20’s to it. I’m just too impatient to wait if I have something cheap on hand that should hopefully work as proof of concept. And I was thinking the fT and power wouldn’t really be a problem for this particular case, so thanks for confirming!
 
GlasgowGrip said:
Okay, so when I run it on the DBT with no outputs installed, I have +/-0.6v on the predriver cans, and they respond to adjustment of the bias pot. It will slowly drift up or down, and if set low enough it will stay constant. Same if I measure the bias the standard way (as if the outputs were installed even though they are not). But I think I may have a problem - if I set bias to the 12-15mV range, it will not remain stable and will slowly start to run away if I don't back off the pot. Is this because the output transistors aren't there to warm up the heat sink? Or do you think it indicates a problem with the bias compensation circuit? I am re-using the original germanium diode CR801 installed on top of the upper predriver Q507 heatsink, but the original Q801 was kaput so I replaced it with an MJE180, hFE = 142. Should I just throw in the outputs and see if that adds enough heat to the bias thermal feedback loop to stabilize it? Yikes.
Click to expand...

You will not be able to measure any bias current across your emitter resistors when there are no outputs, as there is nothing to be powered, same goes for the collector to collector method. I wouldn't sweat the behavior, as without outputs you'd be measuring from one floating point to another floating point, so whatever voltage is there is going to be rather random.
 
GlasgowGrip said:
Okay, so when I run it on the DBT with no outputs installed, I have +/-0.6v on the predriver cans, and they respond to adjustment of the bias pot. It will slowly drift up or down, and if set low enough it will stay constant. Same if I measure the bias the standard way (as if the outputs were installed even though they are not). But I think I may have a problem - if I set bias to the 12-15mV range, it will not remain stable and will slowly start to run away if I don't back off the pot. Is this because the output transistors aren't there to warm up the heat sink? Or do you think it indicates a problem with the bias compensation circuit? I am re-using the original germanium diode CR801 installed on top of the upper predriver Q507 heatsink, but the original Q801 was kaput so I replaced it with an MJE180, hFE = 142. Should I just throw in the outputs and see if that adds enough heat to the bias thermal feedback loop to stabilize it? Yikes.
Click to expand...
To add, the voltage will likely drift a bit as the predrivers heat and cool, they run them HOT in these. Usually when biasing these they need a good 30 minute warmup to be good
 
Have both channels up and running now, both are slightly warm to the touch, but not concerning. With a square wave on an oscilloscope I can see some small damped oscillations/ringing, but nothing majorly concerning since there's nothing increasing in amplitude over time. Bias is stable in both channels, though takes a while to reach an equilibrium point which is normal for the 250/m.


I also modified the relay circuit to accept a modern relay as I felt it was a better bet than modifying the relay itself, as if the relay ever needs replacing again it makes the next techs job easier. Also replaced the rectifier diodes with modern Ultrafast 6A diodes.

Last night I discovered that the left meter (of course I didn't test the left channel) has gone OC, wound up being a simple fix of running a small wire inside the meter to bridge where the continuity was broken. After a ton of fiddling I was able to get both meters 100% (the other had the tension coil slightly disformed).

Amp is currently up and running, seems stable, cranked it for a bit on some 4ohm speakers without issue. I was also able to get ahold of a replacement lamp holder so I could remove the hotglue abomination.

Just need to calibrate the meters and it should be good to go.

(Next post will have parts used and some technical details)
 
So onto my conclusion post I guess, unless the unit nukes itself...
Things to note:
When rebuilding these amps always verify that you have +-.5-.7v on your predriver collector/cans, this is the bias voltage and should adjust with the pot, if not it is not safe to install outputs.

The bias voltage is directly affected by your DC offset in early revisions of this amp. Not entirely sure why but I believe it's because the DC adjustment interacts with your +-.7v. regardless, if you see a huge difference between the +.7 and -.7 then chances are your DC offset is adjusted incorrectly, or your diff pair is too low gain, or potential issues elsewhere.

ALWAYS test the drivers in the Amplifier, the "good channel" in my amp had one with a normal HFE of 60 and the other with a damaged HFE of 13, was just waiting to blow. That being said, predriver matching seems to be much more important than driver matching as this is the 2nd unit I've done that's been stable with one driver at 90 or so HFE and one driver at 170 or so HFE, I think Marantz may have some tolerance on the gain ranges provided.

If you have no bias movement on a 100w bulb with both channels (I ran into this) try a 150 or 200w bulb, the Vdrop from the 100w bulb can be high enough to prevent the amp from biasing.

MJE180G worked just fine for my bias Transistor once I installed the higher HFE diff pair, but ymmv since Marantz made so many mid production changes.

Check your resistors, especially solid carbon ones, they like to drift.

Anyway, as for my parts list:
2N4250 - PN4250
SPS1951 - BC547A
SPS 1952 - BC560C
2N4125 - 2N3906
2N4123 - 2N3904
MPSA20 - Central Semi MPSA20 (Still made)
Differential pair:
SPS1928 - 2N5089 matched with HFEs over 800. Gain is critical here due to how the DC circuit affects bias. BC550C would probably work too, but the pinouts are different.
Predrivers:
(gain match critical between NPN and PNP)
SJ2582 - 2N5415
SJ2583 - 2N3440
Drivers: (MJE parts should work, however you may have to order a few of each P/N of the PNP due to gain)
Using the below parts I was able to reasonably match the requirements set by Marantz. If you use MJE15031 you may wind up with 300+ HFE which wouldn't work, unfortunately it is luck of the draw as they don't bin these parts by gain.
SJ2585 - MJE15033
SJ2586 - MJE15032
Outputs:
SJ2520/2519 - used MJ21195/96 here due to availability, luckily these rarely differ by much in HFE.
MPSA93 or SPS2324 - used modern production MPSA93
MPSA43 or SPS2325 - used NTE287 due to availability
Bias diode replaced with NOS original part. Both did test okay however.
---------------------------------------
Relay board:
Rectifier Diodes:
MR752 - HER605G
1N4003 - UF4004
Relay:
OMRON MY2-DC24 - modified in by swapping pins 13 and 9 as well as 12 and 14. Check that your DC supply is within +-10% of 24V, mine was not (was closer to 30V) so a dropper resistor was added to extend coil life.
Transistors:
SS47 - MJ182G
SPS438 - BC546B

All that's left now is to modify in some better RCA connectors (if I feel like it) and this one should be good to go!

One thing I would possibly consider is modifying in larger predriver heatsinks, with the stock heatsink they run very hot.
 
@goldswimmerb: Congrats on getting your amp running well, and thanks for such a thorough writeup. Well done!

To continue our earlier discussion, I have swapped in a BC547 instead of the MJE180 for the Q801 bias transistor, and everything still behaved the same way without the outputs. My pre-drivers were still running at +/-0.6v, so I went ahead and mounted the outputs (MJ21193/MJ21194) and brought it up on the DBT. All good, bias was now settable and stable. So next I brought it up without the DBT. Set the bias and I'm just letting it idle for a while, with the input shorted (600 ohm resistor). Bias and offset are stable, and everything is staying cool, so next I'll try it with a signal generator and a dummy load and start to see what I really have. But so far so good - thanks again for your help!
 
GlasgowGrip said:
@goldswimmerb: Congrats on getting your amp running well, and thanks for such a thorough writeup. Well done!

To continue our earlier discussion, I have swapped in a BC547 instead of the MJE180 for the Q801 bias transistor, and everything still behaved the same way without the outputs. My pre-drivers were still running at +/-0.6v, so I went ahead and mounted the outputs (MJ21193/MJ21194) and brought it up on the DBT. All good, bias was now settable and stable. So next I brought it up without the DBT. Set the bias and I'm just letting it idle for a while, with the input shorted (600 ohm resistor). Bias and offset are stable, and everything is staying cool, so next I'll try it with a signal generator and a dummy load and start to see what I really have. But so far so good - thanks again for your help!
Click to expand...
I'm not quite sure how much power is disapated by Q801, but BC547 is only rated at 100ma, whereas MJE180 is rated at 3A. Wouldn't consider that an appropriate substitute as if the bias Transistor pops the unit will destroy pretty much everything else.
 
goldswimmerb said:
I'm not quite sure how much power is disapated by Q801, but BC547 is only rated at 100ma, whereas MJE180 is rated at 3A. Wouldn't consider that an appropriate substitute as if the bias Transistor pops the unit will destroy pretty much everything else.
Click to expand...
Agreed, but remember your theory is that Q801 (the "epoxy transistor") was probably an MPSA20, based on the schematic from the Model 32. MPSA20 and BC547 are both small signal TO-92 NPN's with max collector current rating of 0.1A. So if the theory is correct, that means the MJE180 was total overkill as far as current/power ratings.

I've been monitoring it, it runs cool (i.e. no warmer than the heatsink). Also, I plan to replace the BC547 with an MPSA20 as soon as I get some in from Mouser.
 
Last edited:
goldswimmerb said:
So onto my conclusion post I guess, unless the unit nukes itself...
Things to note:
When rebuilding these amps always verify that you have +-.5-.7v on your predriver collector/cans, this is the bias voltage and should adjust with the pot, if not it is not safe to install outputs.

The bias voltage is directly affected by your DC offset in early revisions of this amp. Not entirely sure why but I believe it's because the DC adjustment interacts with your +-.7v. regardless, if you see a huge difference between the +.7 and -.7 then chances are your DC offset is adjusted incorrectly, or your diff pair is too low gain, or potential issues elsewhere.

ALWAYS test the drivers in the Amplifier, the "good channel" in my amp had one with a normal HFE of 60 and the other with a damaged HFE of 13, was just waiting to blow. That being said, predriver matching seems to be much more important than driver matching as this is the 2nd unit I've done that's been stable with one driver at 90 or so HFE and one driver at 170 or so HFE, I think Marantz may have some tolerance on the gain ranges provided.

If you have no bias movement on a 100w bulb with both channels (I ran into this) try a 150 or 200w bulb, the Vdrop from the 100w bulb can be high enough to prevent the amp from biasing.

MJE180G worked just fine for my bias Transistor once I installed the higher HFE diff pair, but ymmv since Marantz made so many mid production changes.

Check your resistors, especially solid carbon ones, they like to drift.

Anyway, as for my parts list:
2N4250 - PN4250
SPS1951 - BC547A
SPS 1952 - BC560C
2N4125 - 2N3906
2N4123 - 2N3904
MPSA20 - Central Semi MPSA20 (Still made)
Differential pair:
SPS1928 - 2N5089 matched with HFEs over 800. Gain is critical here due to how the DC circuit affects bias. BC550C would probably work too, but the pinouts are different.
Predrivers:
(gain match critical between NPN and PNP)
SJ2582 - 2N5415
SJ2583 - 2N3440
Drivers: (MJE parts should work, however you may have to order a few of each P/N of the PNP due to gain)
Using the below parts I was able to reasonably match the requirements set by Marantz. If you use MJE15031 you may wind up with 300+ HFE which wouldn't work, unfortunately it is luck of the draw as they don't bin these parts by gain.
SJ2585 - MJE15033
SJ2586 - MJE15032
Outputs:
SJ2520/2519 - used MJ21195/96 here due to availability, luckily these rarely differ by much in HFE.
MPSA93 or SPS2324 - used modern production MPSA93
MPSA43 or SPS2325 - used NTE287 due to availability
Bias diode replaced with NOS original part. Both did test okay however.
---------------------------------------
Relay board:
Rectifier Diodes:
MR752 - HER605G
1N4003 - UF4004
Relay:
OMRON MY2-DC24 - modified in by swapping pins 13 and 9 as well as 12 and 14. Check that your DC supply is within +-10% of 24V, mine was not (was closer to 30V) so a dropper resistor was added to extend coil life.
Transistors:
SS47 - MJ182G
SPS438 - BC546B

All that's left now is to modify in some better RCA connectors (if I feel like it) and this one should be good to go!

One thing I would possibly consider is modifying in larger predriver heatsinks, with the stock heatsink they run very hot.
Click to expand...
thanks for the info! I was concerned about the predrivers running so hot on mine as well, but looks like that's the way it is. I'm printing up some of the relevant posts in this thread and adding to my service manual for future reference. Thanks to yours, and others info, my 240 is now stable and THD measures as low as my signal generator allows(.07%)
( I need a cleaner audio generator)
 
rich l said:
thanks for the info! I was concerned about the predrivers running so hot on mine as well, but looks like that's the way it is. I'm printing up some of the relevant posts in this thread and adding to my service manual for future reference. Thanks to yours, and others info, my 240 is now stable and THD measures as low as my signal generator allows(.07%)
( I need a cleaner audio generator)
Click to expand...
oooops!For the record I just realized typo above. Should be (.7%), not (.07%).
 
rich l said:
thanks for the info! I was concerned about the predrivers running so hot on mine as well, but looks like that's the way it is. I'm printing up some of the relevant posts in this thread and adding to my service manual for future reference. Thanks to yours, and others info, my 240 is now stable and THD measures as low as my signal generator allows(.07%)
( I need a cleaner audio generator)
Click to expand...
My 240 runs hot too. Interestingly enough the 1200 is cool as a cucumber.
 
ocdecio said:
My 240 runs hot too. Interestingly enough the 1200 is cool as a cucumber.
Click to expand...
I dont know if its my imagination or not, but when i dialed the bias back to 12mV or so the predrivers were a lot cooler than when it was up to 15-18 mV. I left it on the lower end thinking cooler is better. I need to learn how to interpret the output waveform of the distortion analyzer to see what the crossover distortion is. 12mV might be too low for that. Any thoughts on that would be appreciated.
 
rich l said:
I dont know if its my imagination or not, but when i dialed the bias back to 12mV or so the predrivers were a lot cooler than when it was up to 15-18 mV. I left it on the lower end thinking cooler is better. I need to learn how to interpret the output waveform of the distortion analyzer to see what the crossover distortion is. 12mV might be too low for that. Any thoughts on that would be appreciated.
Click to expand...
I don't have a distortion analyzer (in my wishlist!) but crossover distortion is failrly easy to see with a signal gen and a scope. Set it to like 1khz / 100mv and observe the wave as you lower the bias. Where the signal crosses the zero mark there will be a "flat" area. That's the crossover distortion. As you increase bias it disappears.

Of course you may not want to "over-bias" because as you can see it will indeed increase power consumption (for not much better results). I leave mine closer to 12mv too.

image_25679.jpg
 
ocdecio said:
I don't have a distortion analyzer (in my wishlist!) but crossover distortion is failrly easy to see with a signal gen and a scope. Set it to like 1khz / 100mv and observe the wave as you lower the bias. Where the signal crosses the zero mark there will be a "flat" area. That's the crossover distortion. As you increase bias it disappears.

Of course you may not want to "over-bias" because as you can see it will indeed increase power consumption (for not much better results). I leave mine closer to 12mv too.

View attachment 2244041
Click to expand...
I couldnt see any crossover distortion on my scope( 150mhz tektronic) just scoping the output. I was looking for it while running the amp pretty hard. Is it more obvious at a lower signal level?
 
rich l said:
I couldnt see any crossover distortion on my scope( 150mhz tektronic) just scoping the output. I was looking for it while running the amp pretty hard. Is it more obvious at a lower signal level?
Click to expand...
I think so - at lower signal levels the lack of bias is more obvious I think. If you are getting a clear, well centered waveform that's a good sign. I also go for the square wave to check any issues with bass/treble. A good squarewave at 20kHz gives me peace of mind.
 
catrafter said:
IMHO, negative feedback covers up 'visible' crossover distortion in most cases.

Tom
Click to expand...
That's quite interesting - how would one be able to detect/show it? I managed to see it in the 1200 with low bias. But as you say, things are not that simple in a power amp with negative feedback.
 
ocdecio said:
That's quite interesting - how would one be able to detect/show it? I managed to see it in the 1200 with low bias. But as you say, things are not that simple in a power amp with negative feedback.
Click to expand...
Would you look for a similar signature at the output of the distortion analyzer? Its pretty obvious on the screenshot you posted.
 
You must log in or register to reply here.
Back
Top Bottom