Having tested all morning and some of the afternoon at work on crud speakers and source I can confirm the sound is smoother and more layed back. There seems to be a better balance to the sound.
Can't wait for the proper test tonight.
I'm not sure what the VA is on the KI tx but I assume its roughly 250 - 300VA and with doubling up it's around the 500-600VA mark.
Brent
Can't wait for the proper test tonight.
I'm not sure what the VA is on the KI tx but I assume its roughly 250 - 300VA and with doubling up it's around the 500-600VA mark.
Brent
Thomo said:
Hi Lee,
I've only heard his whole system, and not compared amps directly, so it's impossible to say! I do suspect though mine sounds a tad soft but this is likely due to the 100va mains transformers. I might actually do as Brent did and double up as an experiment - I have more of the same transformers still!
Ask me again when I've built a proper one with a serious power supply and all "high end" parts. I plan to do this, and my £900 Roksan Caspian is still amplifying our lounge TV.... so the LM3886 is good enough
Simon
Ok thanks.
I think i'll make one but have to wait and get a tx. The only one i've got has a common ground and i cant seem to get it to work with that.
To be honest I really didn't think that a PM66KI would even come close to LM3886, my ki sounds very dry and way too bright for me. My Quad sounds better in every department.
Maybe it's those little caps of Brent's that help.
Lee.
I think i'll make one but have to wait and get a tx. The only one i've got has a common ground and i cant seem to get it to work with that.
To be honest I really didn't think that a PM66KI would even come close to LM3886, my ki sounds very dry and way too bright for me. My Quad sounds better in every department.
Maybe it's those little caps of Brent's that help.
Lee.
Brent's doesn't sound dry (bit bright though!). All the nice small parts, and the tiny (massive) psu caps and doubling up on the transformer must do a lot of good for it. He's also running it off a Russ Andrews Reference Powerkord, and it sits on a Torlyte platform and has some damping. He will be reporting success with the transformer doubling
I should say though that his system absolutely prefers well recorded and simple music. This is mostly a room and extreme CD player thing though in my opinion. (mine plays rock better)
Simon
I should say though that his system absolutely prefers well recorded and simple music. This is mostly a room and extreme CD player thing though in my opinion. (mine plays rock better)
Simon
The LM3886, im spite of having good power supply noise rejection, is sensitive to the arrangement of psu caps. You will not get the best without some care here. For me 2 x 3,300u per rail with 1R/100nF snubber works very well. Local decoupling is 100u Rubycon ZA or ZL. This should have a different thread as we're not talking about the PM66KI anymore. I will open a thread when I start work on my new amp.
Simon
Simon
Well I tested the amp out for roughly 8 hours last night. It's all positive.
Deeper bass , more musical bass , transients are better , timing of bass is very good, the mid to lower vocals have more weight and depth, the stereo image is a tad wider and what surprised me the most is the extra grip and control in the treble.
The moral of the story is bigger VA is better lol
Brent
Deeper bass , more musical bass , transients are better , timing of bass is very good, the mid to lower vocals have more weight and depth, the stereo image is a tad wider and what surprised me the most is the extra grip and control in the treble.
The moral of the story is bigger VA is better lol
Brent
Well now the amp has been left on a few days its really settled in and the bass depth is superb. I really am enjoying this mod.
It also got me interested in how it now compared to the Marantz PM17 KI amp. I bought the service manual for it to compare. The output section is basically identical, I assume my two TX are better than the nice one in the 17KI and my caps are better. The signal path is shorter in the 66 too. The only thing I can see what is probably better in the 17 is the HDAM pre amp section as apposed to the ic in the 66.
With the tx and cap upgrade my +/- V to the output section has raised by 6V to 44V which is also identical to the PM17.
I do believe my amp will sound better than one of these PM17KI now.
Now lets look what else I could do.....
Brent
It also got me interested in how it now compared to the Marantz PM17 KI amp. I bought the service manual for it to compare. The output section is basically identical, I assume my two TX are better than the nice one in the 17KI and my caps are better. The signal path is shorter in the 66 too. The only thing I can see what is probably better in the 17 is the HDAM pre amp section as apposed to the ic in the 66.
With the tx and cap upgrade my +/- V to the output section has raised by 6V to 44V which is also identical to the PM17.
I do believe my amp will sound better than one of these PM17KI now.
Now lets look what else I could do.....
Brent
I have made my Amp run closer to Pure class A.
I have lifted the idle voltage from 14mV to 28mV. It now has an idle current of 140mA. Also I changed the cheap boggo pots to better 24 turn sealed jobbies for fine adjustment (plus much less chance of drift).
Of course it runs warmer but the power and attack of music is superb, the amp sounds more powerful (which of course it is not). Due to the average listening volumes we listen at and the new idle current I would say I am running a good 90%-95% class A. Also the transistor switch on is greatly reduced too.
Brent
I have lifted the idle voltage from 14mV to 28mV. It now has an idle current of 140mA. Also I changed the cheap boggo pots to better 24 turn sealed jobbies for fine adjustment (plus much less chance of drift).
Of course it runs warmer but the power and attack of music is superb, the amp sounds more powerful (which of course it is not). Due to the average listening volumes we listen at and the new idle current I would say I am running a good 90%-95% class A. Also the transistor switch on is greatly reduced too.
Brent
Here is the mods list (well the bits I can remember)
PSU
• FITTED EXTRA KI TX (NOW RUNNING 2) VA approx 500VA
• REPLACED BRIDGE RECTIFIER (D801) WITH 4X MBR20 20A SCHOTTKY DIODES
• REPLACED C801 C802 WITH 56V 22000uF SILMICS
• REPLACED C717 C718 C719 WITH 50V 1000uF BLACK GATE STD TYPE
• MOVED Q801 Q802 ONTO HEATSINK
• FITTED 100nF , 10nF AND 1 OHM 3W RES ACROSS C801 C803
INPUT
• FITTED QUALITY RCA TO CD INPUT
• REPLACED C701 C702 WITH 50V 10uF BLACK GATE AC TYPE
• REPLACED C705 C706 WITH 16V 100uF BLACK GATE N TYPE
• REPLACED C709 C710 WITH 100pF SILVER MICA 1% TOL
OUTPUT
• REPLACED C751 C752 WITH 25V 100uF ELNA CEREFINE (ON BIAS CIRCUIT)
• REPLACED R755 R756 WITH 2K 24 TURN PRECISION POT (ON BIAS CIRCUIT)
• REPLACED C753 C754 C755 C756 WITH 120pF SILVER MICA 1% TOL
• REPLACED INTERNAL OUTPUT WIRE WITH CABLE TALK MONITOR 2
• ADJUSTED OUTPUT IDLE FROM 14mV (70mA) to 20mV (100mA)
OTHER
• FITTED CHASSIS AND LID WITH DAMPING MATERIAL
• UNIT SITS ON OAK CONE FEET
Brent
PSU
• FITTED EXTRA KI TX (NOW RUNNING 2) VA approx 500VA
• REPLACED BRIDGE RECTIFIER (D801) WITH 4X MBR20 20A SCHOTTKY DIODES
• REPLACED C801 C802 WITH 56V 22000uF SILMICS
• REPLACED C717 C718 C719 WITH 50V 1000uF BLACK GATE STD TYPE
• MOVED Q801 Q802 ONTO HEATSINK
• FITTED 100nF , 10nF AND 1 OHM 3W RES ACROSS C801 C803
INPUT
• FITTED QUALITY RCA TO CD INPUT
• REPLACED C701 C702 WITH 50V 10uF BLACK GATE AC TYPE
• REPLACED C705 C706 WITH 16V 100uF BLACK GATE N TYPE
• REPLACED C709 C710 WITH 100pF SILVER MICA 1% TOL
OUTPUT
• REPLACED C751 C752 WITH 25V 100uF ELNA CEREFINE (ON BIAS CIRCUIT)
• REPLACED R755 R756 WITH 2K 24 TURN PRECISION POT (ON BIAS CIRCUIT)
• REPLACED C753 C754 C755 C756 WITH 120pF SILVER MICA 1% TOL
• REPLACED INTERNAL OUTPUT WIRE WITH CABLE TALK MONITOR 2
• ADJUSTED OUTPUT IDLE FROM 14mV (70mA) to 20mV (100mA)
OTHER
• FITTED CHASSIS AND LID WITH DAMPING MATERIAL
• UNIT SITS ON OAK CONE FEET
Brent
Hi Brent,Simont and all,
I interested with your marantz PM66 modded and get inspiration to upgrade my yamaha RX-V490 power amp.
1. I replaced caps (PSU) 2x6.800 uF with new Black elna 2x 10.000 and by pass with wima 3.3 uF and 100 nF.
2. Unfortunatelly I don't have manual schematic so I don't know what is the next step to do it?
Regards
aquar
I interested with your marantz PM66 modded and get inspiration to upgrade my yamaha RX-V490 power amp.
1. I replaced caps (PSU) 2x6.800 uF with new Black elna 2x 10.000 and by pass with wima 3.3 uF and 100 nF.
2. Unfortunatelly I don't have manual schematic so I don't know what is the next step to do it?
Regards
aquar
rowemeister said:
Hi Brent,
I decided to do some measurements and opened up my 66KI last night.
I get DC offsets of -12mV/-12.4mV which is good, presumably down to reasonable matching of the input pair as I can't see any offset adjust.
I measured the voltage across the emitter resistors (to check idle current) and got 14mV/18mV. This was after about 10 minutes. I tweaked the left channel up to 18mV and I haven't listened yet.
I have a couple of questions.
It seems to drift about more on the right channel... should it be stable? It seems to float about between 15mV and 18mV. This is with the lid off... which brings me to another question. If the bias current goes up with heat, then this reading would surely be much higher if the amp has its lid on, and was in a rack.
Also, if more heat creates more bias and more bias creates more heat is there not a danger of some kind of thermal runaway?
So what is the best way to measure this?
I have the PM-6010SE service manual (thanks Thommo!) which has the same output stage and that suggests adjusting for 10mV which is only 50mA of bias current. This seems low but are they accounting for the fact that this will go up in normal operation I wonder.
My other question is regarding the main PSU caps. The 12,000uF boys. On mine they are slightly domed at the top. It doesn't look like the vents have ruptured so is this normal?
Questions, questions!
Glenn
Hi Glenn
They all do that on the Caps! its normal. Nothing to worry about.
Idle... idle current drops when very hot so thats not an issue.
You can only adjust the idle current from cold to accuratly set it up. Adjusting one channel when warm by 4mV will be very different from cold.
First of all change the pots to 2K 24 turn jobbies. These dont cause drift and make setting up more accurate. The new pots are set to the 1K position and are not far off the mark.
Then follow this procedure (i'm upping your mA to 100mA BTW)
Get two DVM if you can and monitor both channels together.
From cold
30 sec to 1 min - adjust both channals to 6mV
1 min to 2 mins - adj to 12mV
2 min to 4 min - adj to 16mV
6 min + adj to 20mV
leave on for 30 mins and check for running. Run quite loud for a bit (with DVM still wired on). Turn vol down to min and look at mV. It will drop low and slowly rise and settle. Check both channels are pretty much the same to settle.
Now turn off for 15-20 mins allowing to cool.
Turn on and watch the mV rise on both channels, they should match each other pretty well all the way to the 20mV mark. If they settle to say 18mV or 22mV thats fine.
If they settle at say 26ish mV each you may find it gets hot in there and may want to redo the procedure, but they should settle to what you set them too.
Try to get as much ventilation as possible, I have to have mine on top of the rack as my air gap is very small.
If further down the line all is superbly well with heat etc (20mV is fine) you may want to raise it slightly.
Brent
They all do that on the Caps! its normal. Nothing to worry about.
Idle... idle current drops when very hot so thats not an issue.
You can only adjust the idle current from cold to accuratly set it up. Adjusting one channel when warm by 4mV will be very different from cold.
First of all change the pots to 2K 24 turn jobbies. These dont cause drift and make setting up more accurate. The new pots are set to the 1K position and are not far off the mark.
Then follow this procedure (i'm upping your mA to 100mA BTW)
Get two DVM if you can and monitor both channels together.
From cold
30 sec to 1 min - adjust both channals to 6mV
1 min to 2 mins - adj to 12mV
2 min to 4 min - adj to 16mV
6 min + adj to 20mV
leave on for 30 mins and check for running. Run quite loud for a bit (with DVM still wired on). Turn vol down to min and look at mV. It will drop low and slowly rise and settle. Check both channels are pretty much the same to settle.
Now turn off for 15-20 mins allowing to cool.
Turn on and watch the mV rise on both channels, they should match each other pretty well all the way to the 20mV mark. If they settle to say 18mV or 22mV thats fine.
If they settle at say 26ish mV each you may find it gets hot in there and may want to redo the procedure, but they should settle to what you set them too.
Try to get as much ventilation as possible, I have to have mine on top of the rack as my air gap is very small.
If further down the line all is superbly well with heat etc (20mV is fine) you may want to raise it slightly.
Brent
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.