Different Vintages?
I found a schematic online of PL400. From the look of this one, Zin would be about 12700 Ohms. Does yours have much higher input resistors? If so, was it a mod, or just a different vintage?
OK...here's a hint, the website:
Phase Linear PL-400 | Owners Manual, Service Manual, Schematics, Free Download | HiFi Engine
said:
Input impedance: PL400C-47 kOhms, PL14A-10 kOhms, PL14B-39 kOhms
I'm thinking that I may not have all the details, since there's no input blocking cap, but maybe this was just an amp for the brave.
Anyway, if you change R3, the only issue will be a bit of offset voltage at the output. If this schematic is correct, the diff pair transistors each have input bias current of around 75/200000/2/150=1.25 micro-Amps. If the input were AC coupled, then the output offset due to input bias current would be about (R3-22000)*1.25e-6.
If you choose R3=100K, the speaker offset voltage would change from its present value (assumed due largely to offset voltage) by 97.5 mV. That's probably tolerable.
I found a schematic online of PL400. From the look of this one, Zin would be about 12700 Ohms. Does yours have much higher input resistors? If so, was it a mod, or just a different vintage?
OK...here's a hint, the website:
Phase Linear PL-400 | Owners Manual, Service Manual, Schematics, Free Download | HiFi Engine
said:
Input impedance: PL400C-47 kOhms, PL14A-10 kOhms, PL14B-39 kOhms
I'm thinking that I may not have all the details, since there's no input blocking cap, but maybe this was just an amp for the brave.
Anyway, if you change R3, the only issue will be a bit of offset voltage at the output. If this schematic is correct, the diff pair transistors each have input bias current of around 75/200000/2/150=1.25 micro-Amps. If the input were AC coupled, then the output offset due to input bias current would be about (R3-22000)*1.25e-6.
If you choose R3=100K, the speaker offset voltage would change from its present value (assumed due largely to offset voltage) by 97.5 mV. That's probably tolerable.
http://www.dvq.com/hifi/images/phase400.pdf
Change R2 to whatever you want, adjust R8 for minimum DC off-set.
Change R2 to whatever you want, adjust R8 for minimum DC off-set.
My input board is the PL-14, not the one in that schematic.
The PL 14 is far different than that... it uses a second differential pair; there is no 100K pot like R8 on the board.
I'm thinking that the input resistor (R1 = 10K) that is in the signal path to the first differential pair Q1 and Q2, and the resistor that sets input impedance (R3 = 39K) form a voltage divider.
So if I want to increase the impedance of R3 by 1/2, then I should increase R1 by 1/2 as well. So with R3 at 60K means, R1 should be about 15K.
I'm assuming that 1/4 watt resistors ought to be good enough.
thanks for the hifiengine link...I checked the schematics there in the service manual that seems like yours...my earlier calculations should be ok...it's close enough...if you make R1 go from 10K to 15K and R3 go from 39K to 60K, everything should still work fine. As you suspected, 1/4 Watt resistors are plenty big enough.
Good luck...and be careful...
Good luck...and be careful...
Thanks, for checking! I don't understand your calculation, but assuming that it is correct, 39K would imply a 21.3 mv offset.
OTOH, according to the manual, the offset is a function of how well Q1 and Q2 are matched.
Others here have suggested that the offset could also be decreased by replacing all the old electrolytics on the PL-14 board.
Is there a way that through component selection of the resistors on the collectors of Q1/Q2 or other parts that the offset can be minimized? As I understand it, the lower the offset, the lower the cross-over distortion....
OTOH, according to the manual, the offset is a function of how well Q1 and Q2 are matched.
Others here have suggested that the offset could also be decreased by replacing all the old electrolytics on the PL-14 board.
Is there a way that through component selection of the resistors on the collectors of Q1/Q2 or other parts that the offset can be minimized? As I understand it, the lower the offset, the lower the cross-over distortion....
Excellent! Yes, it is at +75mv, and those transistors are NPN.
I'm guessing that making R3 100K (with corresponsing increase to R1 = 25K) will decrease that offset even more.
I might even get a chance to try it tonight. I will post my results here after.
Thank you so much for the help!
Now I am very keen to check it out, all the resistors and the 220pF cap are on the input RCA's. It should be simple. At least it will be easier than replacing the power supply caps with 4-pole electrolytics...
Wish me luck!
I'm guessing that making R3 100K (with corresponsing increase to R1 = 25K) will decrease that offset even more.
I might even get a chance to try it tonight. I will post my results here after.
Thank you so much for the help!
Now I am very keen to check it out, all the resistors and the 220pF cap are on the input RCA's. It should be simple. At least it will be easier than replacing the power supply caps with 4-pole electrolytics...
Wish me luck!
As I understand it,
If the DC offset in the differential pai4 Q1,Q2 is related to the difference between [collector current/beta(Q1)/input resistor] and [collector current/beta(Q2)/input resistor] then by doubling the input resistor value, the contribution to the offset by the different betas in the differential pair should decrease by 1/2, and going 2.5 times should decrease to 40% of the original value -- assuming that the collector current is equal.
The corollary is that if the offset behaves in this manner, then the electrolytics are fine. If the offset is not changed much, then the electrolytics, not the mismatched betas, are the primary cause for the DC offset.
If the DC offset in the differential pai4 Q1,Q2 is related to the difference between [collector current/beta(Q1)/input resistor] and [collector current/beta(Q2)/input resistor] then by doubling the input resistor value, the contribution to the offset by the different betas in the differential pair should decrease by 1/2, and going 2.5 times should decrease to 40% of the original value -- assuming that the collector current is equal.
The corollary is that if the offset behaves in this manner, then the electrolytics are fine. If the offset is not changed much, then the electrolytics, not the mismatched betas, are the primary cause for the DC offset.
doesn't really matter. It will make a slight difference in the gain of the channel, so if you want to be really careful about stereo balance, you'll modify both channels the same way.
Note that you will get a bit lower corner on the lowpass filter when you go from 10K:39K to 15K:60K...There's a 220 pF cap, which makes input filter at
1/(2*pi*(10K||39K)*220e-12)=90.89 kHz to 1.5X lower, or about 60.6 kHz...That's still so far above the audio band that it should not matter.
Note that you will get a bit lower corner on the lowpass filter when you go from 10K:39K to 15K:60K...There's a 220 pF cap, which makes input filter at
1/(2*pi*(10K||39K)*220e-12)=90.89 kHz to 1.5X lower, or about 60.6 kHz...That's still so far above the audio band that it should not matter.
Well, I made a bunch of changes today.
Changed power cord to ecosse "bug purple" - it's 14 gauge, and just fit in the strain relief -- with some help with a drill.
Then, completed the 4-pole cap installation -- I had used spades and slipped them over the solder lugs. They are all now soldered in place.
Swapped resistors with 14.8K and 62K (matched with my DMM). [ Sonically, this lifted a veil ]
I was on a roll, so I swapped out the electrolytic caps on the driver board. I replaced 2 nichicon, 6 cel and 2 rayrex. The cel were undersized, 47 uF @ 35V, where the schematic says 47 uF @ 40V. (installed is 47uF @50V) . Every cap I installed is MUCH smaller than the ones that were in there.....but they are no-name product.
This was not good. The amp has lost bass and is now bright sounding.
Do electrolytics need to break in or are there specific products that are considered better for the Phase?
At least I don't fear poking around in the phase....
Oh, I did rebias to .442 V, under the assumption that higher is better.
The DC offset after changing the input resistors dropped 10mv each channel.
Thanks!
Changed power cord to ecosse "bug purple" - it's 14 gauge, and just fit in the strain relief -- with some help with a drill.
Then, completed the 4-pole cap installation -- I had used spades and slipped them over the solder lugs. They are all now soldered in place.
Swapped resistors with 14.8K and 62K (matched with my DMM). [ Sonically, this lifted a veil ]
I was on a roll, so I swapped out the electrolytic caps on the driver board. I replaced 2 nichicon, 6 cel and 2 rayrex. The cel were undersized, 47 uF @ 35V, where the schematic says 47 uF @ 40V. (installed is 47uF @50V) . Every cap I installed is MUCH smaller than the ones that were in there.....but they are no-name product.
This was not good. The amp has lost bass and is now bright sounding.
Do electrolytics need to break in or are there specific products that are considered better for the Phase?
At least I don't fear poking around in the phase....
Oh, I did rebias to .442 V, under the assumption that higher is better.
The DC offset after changing the input resistors dropped 10mv each channel.
Thanks!
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.