mjraudio said:...Could I decrease the capacitance value of C101 and change the bandwidth and frequency response of the amplifier? Ideally I would want as wide a bandwidth as possible, so that I can take full advantage of the 1Hz-100kHz frequency range of SACD (when I buy my Sony SCD-1 SACD player someday$$$
The 555i has a 1uF input cap bandwidth is 9 or 10Hz, yes u can change to vary the bandwidth... however what bandwidth does your speaker give you? Probably 35Hz and above -3db.... just guessing. Soon you see 1Hz is over kill.... if so go capless DC-700kHz is a more sexy number than 0.7 to 700,
Any value from 0.8uF to 20uF is ok as long as it fits fine...
Hi Arif,
I agree.
Hi mjraudio,
Leave the original cap in there. It's a really good unit and is overkill. In this case the cap is there to block DC and also roll subsonic frequencies off. Any information below the cutoff frequency of your speakers is wasted power and causes distortion.
The mixture is 10 parts water to 1 part Simple Green.
Always remember that when an enginering team designs an amplifier, the sales department generally pushes the design away from good engineering practices. Then the accountants want to restrict the budget. In general, a good product should not be changed without very good reasons. Cheap stuff, who knows!
-Chris
I agree.
Hi mjraudio,
Leave the original cap in there. It's a really good unit and is overkill. In this case the cap is there to block DC and also roll subsonic frequencies off. Any information below the cutoff frequency of your speakers is wasted power and causes distortion.
The mixture is 10 parts water to 1 part Simple Green.
Always remember that when an enginering team designs an amplifier, the sales department generally pushes the design away from good engineering practices. Then the accountants want to restrict the budget. In general, a good product should not be changed without very good reasons. Cheap stuff, who knows!
-Chris
Thanks for your help Arif & Chris.
I just measured the two resistors that are part of the input DC blocking/high pass filter and they both sum to 51,000ohms resistance.
I calculated the high pass filter frequency and assuming that that little 50VDC / 5.0uF capacitor is exactly 5.0uF, then the bandwidth delivered to the amplifier would be exactly 0.6241Hz-xxxkHz.
5.0uF is definitely overkill, you are both correct!
No I am thinking the opposite way, can I decrease the capacitance of the capacitor to cutoff the bandwidth at a higher frequency, my speakers cannot deliver below 20hZ, so I could probably cut off the input at around 15Hz. Since the capacitor will be tiny to do this, do you see this as a problem? I believe it would significantly reduce system distortion since the amplifier would be working with a bandwidth starting at around 15Hz (if that is the value I go with).
EDIT: Hmm, according to my calculations, I would need an ultra-tiny capacitor the size of an ant. Something is wrong I think...
I just measured the two resistors that are part of the input DC blocking/high pass filter and they both sum to 51,000ohms resistance.
I calculated the high pass filter frequency and assuming that that little 50VDC / 5.0uF capacitor is exactly 5.0uF, then the bandwidth delivered to the amplifier would be exactly 0.6241Hz-xxxkHz.
5.0uF is definitely overkill, you are both correct!
No I am thinking the opposite way, can I decrease the capacitance of the capacitor to cutoff the bandwidth at a higher frequency, my speakers cannot deliver below 20hZ, so I could probably cut off the input at around 15Hz. Since the capacitor will be tiny to do this, do you see this as a problem? I believe it would significantly reduce system distortion since the amplifier would be working with a bandwidth starting at around 15Hz (if that is the value I go with).
EDIT: Hmm, according to my calculations, I would need an ultra-tiny capacitor the size of an ant. Something is wrong I think...
Hmm reading back on my last post...my question to you is:
will limited the bandwidth of the input present an easier load to the amplifier and/or decrease the THD of the amplified output signal?
In other words, since I am not using 1-19Hz, can I just cut those freqs off or is it more complicated than that?
I don't mean to beat this dead horse but I want to understand this.
will limited the bandwidth of the input present an easier load to the amplifier and/or decrease the THD of the amplified output signal?
In other words, since I am not using 1-19Hz, can I just cut those freqs off or is it more complicated than that?
I don't mean to beat this dead horse but I want to understand this.
Hi mjraudio,
If you decrease the input capacitor you will increase the impedance seen by the input stage. Not a good thing.
Use an active crossover to make response adjustments.
-Chris
If you decrease the input capacitor you will increase the impedance seen by the input stage. Not a good thing.
Use an active crossover to make response adjustments.
-Chris
Ahh, so it is much more complicated, I see...the service manual specifies that the input impedance of the 565 is 50,000ohms (1000 ohms less then the input resistance)...This is something that, if changed, I suspect would dramatically alter the amp's performance and/or function. You are right, thanks Chris 🙂
For your viewing pleasure, here are pics of my completely disassembled 565!
I decided to do some Spring Cleaning.
PIC 1, 565 insides are gone!
PIC 2, 565 toroid!
I want to eliminate the "buzz" that is prevalent in the toroid when the amp runs. How can I do so? I tried taking a small strip of the covering on the toroid off. Once it comes off, it can't go back on.
PIC 3, corrosion on 565 toroid!
I'm also rewiring the entire amp (everything that is possible) with the same guage wiring, but MILSPEC (Silver coated copper stranded).
And I'm going to line the internal planes with EMI/RFI shielding foil.
One quick question: do you guys see any potential problems with using an EMI filter at the input? I am going to install an IEC receptacle on the back, and I am thinking of adding on an EMI filter/IEC receptacle, will it help? Can it hurt?
For your viewing pleasure, here are pics of my completely disassembled 565!
I decided to do some Spring Cleaning.
PIC 1, 565 insides are gone!
PIC 2, 565 toroid!
I want to eliminate the "buzz" that is prevalent in the toroid when the amp runs. How can I do so? I tried taking a small strip of the covering on the toroid off. Once it comes off, it can't go back on.
PIC 3, corrosion on 565 toroid!
I'm also rewiring the entire amp (everything that is possible) with the same guage wiring, but MILSPEC (Silver coated copper stranded).
And I'm going to line the internal planes with EMI/RFI shielding foil.
One quick question: do you guys see any potential problems with using an EMI filter at the input? I am going to install an IEC receptacle on the back, and I am thinking of adding on an EMI filter/IEC receptacle, will it help? Can it hurt?
I just saw an IEC cord on ebay for less than $10 with RF shielding/ Ferrite bead... look it up.
Reducing the cap size may actually increase distortion (read up on Self's tests) but definitively, I'd test out and conduct hearing tests if you are hell bent on playing with it. Active crossover shifts the distortion away from the power amp ... but adds it nonetheless.
I have 2 565's, both have Buzzing toroids... unless I reduce the AC to about 110v then they quieten down... 😉
Reducing the cap size may actually increase distortion (read up on Self's tests) but definitively, I'd test out and conduct hearing tests if you are hell bent on playing with it. Active crossover shifts the distortion away from the power amp ... but adds it nonetheless.
I have 2 565's, both have Buzzing toroids... unless I reduce the AC to about 110v then they quieten down... 😉
Hi mjraudio,
High impedance sources with bipolar input transistors does tend to increase distortion. The signal actually distorts before it even gets into the amp. Some manufacturers have used a buffer circuit first (Marantz, Nakamichi), others went to FETs for the diff pair. the buffer will drive the bipolar diff pair with a low impedance so the signal distortion is reduced. Using a smaller coupling cap raises the impedance and increases this effect.
-Chris
It looks like an old fluid spill to me.
Some toriods do that. There are times you can block an impressed DC across your AC line that will cause core saturation (and buzzzzzzzz). There was at least one thread on this. The "cure" was to use some back to back caps and some parallel diodes. That's if I remember correctly.
I hope you stuck it in the sink and washed it out. It'll look like new, well the remaining paint will anyway. 😀
High impedance sources with bipolar input transistors does tend to increase distortion. The signal actually distorts before it even gets into the amp. Some manufacturers have used a buffer circuit first (Marantz, Nakamichi), others went to FETs for the diff pair. the buffer will drive the bipolar diff pair with a low impedance so the signal distortion is reduced. Using a smaller coupling cap raises the impedance and increases this effect.
Resistance. Always go for minimum resistance for an amplifier's AC supply. Many RFI filters are too resistive and you end up with voltage drops with each current spike drawn.
-Chris
Thanks Arif and Chris, I have decided not to go with an EMI filter but just an IEC receptacle on my amps.
Regarding the toroids, can you guys help identify what the specifications are for the 565 toroids?
The only thing that is stamped on the toroid is:
"23-2042-0-0
03 GT"
The toroid can handle 100-240VAC, when running in the 565 on 120VAC, through the KPBC3504 (400V/35A) bridge rectifier it delivers 86.5VDC into the 100V / 35,000uF filter capacitors.
It measures 6.25" in diameter by 2.85" tall.
The primaries are presumably 115/240V 50/60Hz.
How many secondaries are there and what are their voltage and ampere ratings?
It would be great to have this information posted here for everyone to know (myself included).
Thank you! 🙂
Regarding the toroids, can you guys help identify what the specifications are for the 565 toroids?
The only thing that is stamped on the toroid is:
"23-2042-0-0
03 GT"
The toroid can handle 100-240VAC, when running in the 565 on 120VAC, through the KPBC3504 (400V/35A) bridge rectifier it delivers 86.5VDC into the 100V / 35,000uF filter capacitors.
It measures 6.25" in diameter by 2.85" tall.
The primaries are presumably 115/240V 50/60Hz.
How many secondaries are there and what are their voltage and ampere ratings?
It would be great to have this information posted here for everyone to know (myself included).
Thank you! 🙂
It is a 1.2kVA unit, VERY potent. Perhaps 60-0-60vac secondaries.
Rest should be in the Schematic. Unless you plan to run 2 ohm loads, a 600va to 800va would work fine too, though you will lose some power... maybe 30 watts or so. But since the 565 clips around 380 watts into 8 ohms, you still get 350 watts into 8 ohms.
I have substituted the Avel 800va 60-0-60 toroid (Parts Express) and got 375 watts at clip. It will raise the rails bu 1.5 to 2.5vdc but that should not be an issue. They are dead quiet.
I bought 10 of them at less than $70 each.... now they are $100 a piece.
Rest should be in the Schematic. Unless you plan to run 2 ohm loads, a 600va to 800va would work fine too, though you will lose some power... maybe 30 watts or so. But since the 565 clips around 380 watts into 8 ohms, you still get 350 watts into 8 ohms.
I have substituted the Avel 800va 60-0-60 toroid (Parts Express) and got 375 watts at clip. It will raise the rails bu 1.5 to 2.5vdc but that should not be an issue. They are dead quiet.
I bought 10 of them at less than $70 each.... now they are $100 a piece.
Hi Arif,
I haven't seen a single damaged Adcom power transformer in an amp. Tuner / pre - yes. Are you replacing them because of noise issues?
BTW, $100 USD for the main power transformer is pretty cheap.
-Chris
I haven't seen a single damaged Adcom power transformer in an amp. Tuner / pre - yes. Are you replacing them because of noise issues?
BTW, $100 USD for the main power transformer is pretty cheap.
-Chris
Actually I wanted to get 400 watts into 8 ohms at clipping... I was limited to 375 with the 800va unit but did finally get 430watts with the 1kVA unit at clip into 8 ohms with a variac holding my AC to 120vac (it dropped to 114vac in my previous test). But the rails were +/-86.5 vdc with the Avel as opposed +/-83vdc with the stock. The Amp survived 100vdc rails in one of my tests, though I did not have the b@lls to play it loud...
In the end, I went back to the stock... wasn't worth the increase with all the trouble...
I used the 10 donuts in other projects for 555's though 🙂
In the end, I went back to the stock... wasn't worth the increase with all the trouble...
I used the 10 donuts in other projects for 555's though 🙂
anatech said:
Chris: With shipping you are looking at $105 to $108
http://www.partsexpress.com/pe/showdetl.cfm?Partnumber=122-720
2 years ago I remember buying these for $65 a pop !!!
K-amps said:
Arif and Chris,
The voltage in the filter caps is 86.5V, so if the 565 toroid has 60VAC secondaries, then the primaries must be what are run to the bridge rectifier and then into the capacitors (where there is 86.5VDC). Where do the secondaries go?
Also, do all GFA-565 amps have the blue-twin 22AWG wires coming out of the toroid? On my amps, this cable goes into the fan psu pcb on my optional top fan cover. I wonder what voltage comes through that blue wire. Is it just as simple as hooking up a multimeter to the toroid while running? Is there a chance I could damage it?
1.2kVA is an extreme load capacity for the size of the 565 toroid, which is the average size of units with half the current capacity, I'm impressed with Adcom yet again. 🙂
I am not planning on replacing the original Adcom units, but I just wanted to have this information on here. Thanks guys.
For your reference, this toroid looks exceptionally well-built:
DIYCable.com 1KVA toroidal transformer
Arif, I see that you ignored this sign
next to the toroidal transformer part # in the service manual, hehe. 
I would presume that changing the rails even slightly in the amplifier could potentially cause massive problems. Consider the effects of altering the voltage input into the input pcb board, this could cause DC offset issues: problems which are already inherent in this dc-coupled design.
Yes thats a sweet looking 1kva though I doubt u can get it in every voltage u like.
I know the 555ii also has the blue wires for the fan, yes you can measure the voltage from the blue wires to see what u need. The Avel will not have this winding so that will be a separate toroid or E+! for the fan supplies.
The Toroid's Primaries are probably (check the SM) dual 120vac wired in parallel. Hooked to your IEC socket that goes to the mains.
The secondaries are 60-0-60vac. These feed the Bridges.
60 * 1.414 = 85.5 v -1.5v (for the forward bias of the Bridge) leaves you with +/-84vdc rails.
If you are getting 86.5, then you mains is over 120vac... more like 124-125vac.
With 125vac, the Avel (rated for 115vac) will become a 65-0-65vac unit with rails of about +/-90 vdc. This is too high for the pre-drivers which will go past 70c with the extra 6 volts. Long term this is bad, short term no worries.
As your Home Mains is already 125v or so, You are already on the high side of the rails that the amp should be run at... I would leave your unit as is... not worth the trouble or effort to redesign... however if the Buzz is driving you crazy, get a 55-0-55 1kVA or 1.5kVA unit. You will still get over 350 watts at your home voltage...
The Adcom 565 is perhaps the most impressive of Adcoms... the rest are only "nice"... 😉
I know the 555ii also has the blue wires for the fan, yes you can measure the voltage from the blue wires to see what u need. The Avel will not have this winding so that will be a separate toroid or E+! for the fan supplies.
The Toroid's Primaries are probably (check the SM) dual 120vac wired in parallel. Hooked to your IEC socket that goes to the mains.
The secondaries are 60-0-60vac. These feed the Bridges.
60 * 1.414 = 85.5 v -1.5v (for the forward bias of the Bridge) leaves you with +/-84vdc rails.
If you are getting 86.5, then you mains is over 120vac... more like 124-125vac.
With 125vac, the Avel (rated for 115vac) will become a 65-0-65vac unit with rails of about +/-90 vdc. This is too high for the pre-drivers which will go past 70c with the extra 6 volts. Long term this is bad, short term no worries.
As your Home Mains is already 125v or so, You are already on the high side of the rails that the amp should be run at... I would leave your unit as is... not worth the trouble or effort to redesign... however if the Buzz is driving you crazy, get a 55-0-55 1kVA or 1.5kVA unit. You will still get over 350 watts at your home voltage...
The Adcom 565 is perhaps the most impressive of Adcoms... the rest are only "nice"... 😉
K-amps said:
Arif, where does the 1.414 multiplication factor come from?
I just measured my mains power and it is at 123.8VAC...when I have both amplifiers running (on different circuits) it may drop to 119VAC, but my predriver transistors still run at 70C! I really want to decrease the predriver transistor temperature, but unfortunately the input PCB does not allow any space for a larger TO-220 heatsink. What do you think the source of the problem is?
Yes you are right, it is too much trouble to replace the toroid and rebuild the power supply when the 565 toroid is already such an incredible unit. I wonder if I can send it to get "serviced..." Maybe a toroid repair shop can epoxy the windings and reseal it for a small charge...
Kudos, I agree completely. I owned a GFA-545 II before this (still have it on my secondary system), but I was very attracted to the concept of having a dedicated chassis, power supply, and amplifier for each channel, so I upgraded to the 565s and so far am loving it!
Hi Arif,
-Chris
Yes, but I was considering what most distributors of the equipment would charge. Probably around $250 ~300 USD as a guess.
The AC voltage is an RMS measurement. Rectification and filtering act as a peak detector. So the actual eq. would be [(sec VAC) -1.2V] * 1.414 = (DC supply voltage). Of course the forward diode drops will vary with load.
-Chris
Don't dismiss the 545ii, I Like the mids and highs of the 545ii better that the 555 & 565 (though limited in power) but the bass on the 565 is just awesome. As is the power it can deliver!
With 123.8vac... you may be hitting over 400 watts into 8 ohms at clipping. I got 383 watts with 113.5vac.
I don't know if revarnishing or re-epoxying will do a lotta good, but give it a try...
Don't worry about the 70c... they are designed to work that way. They will live even at 90c because of the constant nature of the current passing though.
Also... many have experienced this, the hotter a tranny runs the better it sounds. Ofcourse you get to a point of diminishing and then negative returns after some point.
One mod you can do is to bypass the Large caps with 1000uF then 100uF then 5uF then 0.5uF caps.
Also you can install distributed caps near the OP devices of 1000uF to 2000uF.
These are easy mods and can clear up the mids. But do not over-do it. Also be careful of mechanical reliability in installing these extra caps. Enjoy!!!
With 123.8vac... you may be hitting over 400 watts into 8 ohms at clipping. I got 383 watts with 113.5vac.
I don't know if revarnishing or re-epoxying will do a lotta good, but give it a try...
Don't worry about the 70c... they are designed to work that way. They will live even at 90c because of the constant nature of the current passing though.
Also... many have experienced this, the hotter a tranny runs the better it sounds. Ofcourse you get to a point of diminishing and then negative returns after some point.
One mod you can do is to bypass the Large caps with 1000uF then 100uF then 5uF then 0.5uF caps.
Also you can install distributed caps near the OP devices of 1000uF to 2000uF.
These are easy mods and can clear up the mids. But do not over-do it. Also be careful of mechanical reliability in installing these extra caps. Enjoy!!!
Arif & Chris,
Thanks for the information.
Arif, I don't quite understand what you mean here. On the input stage PCB, there are two types of electrolytic capacitors: 100V/100uF (2) and 25V/220uF (3). Are those the caps that would be bypassed in this modification? Where would the 1000uF caps be installed?
One thing I have wanted to do is increase the capacitance of the two small electrolytic capacitors on the output stage PCBs (two caps per amp)...they are the 100V/47uF ones...The main reason I want to change these is because that value capacitor comes in a reasonably small package and the PCB has holes which are wider than the physical packaging of the electrolytic capacitor of that value...
At the same time, the capacitors used in the 565 are for DC coupling, is it dangerous to change cap sizes when DC offset is such a big concern?
BTW: The 545II is a good unit and is serving me well!
Thanks for the information.
K-amps said:
Arif, I don't quite understand what you mean here. On the input stage PCB, there are two types of electrolytic capacitors: 100V/100uF (2) and 25V/220uF (3). Are those the caps that would be bypassed in this modification? Where would the 1000uF caps be installed?
One thing I have wanted to do is increase the capacitance of the two small electrolytic capacitors on the output stage PCBs (two caps per amp)...they are the 100V/47uF ones...The main reason I want to change these is because that value capacitor comes in a reasonably small package and the PCB has holes which are wider than the physical packaging of the electrolytic capacitor of that value...
At the same time, the capacitors used in the 565 are for DC coupling, is it dangerous to change cap sizes when DC offset is such a big concern?
BTW: The 545II is a good unit and is serving me well!
Heavens no... do not add those caps to the input stages.... 
I meant the main caps can be bypassed.
35,000uF +
1000uF +
100uF +
5uF +
0.5uF +
0.1uF
The first (35kuF) and last one (0.1uF) are already part of the stock set-up. You just need the other 4. Make sure they are 100vdc types and orient them correctly...
You need 1 per rail or 8 caps total...
The other ones will go to the rails that feed the OP devices. .. these are not that important as bypassing the main caps.
Also life will go on without these bypass caps, I just mentioned it after I saw you dis-assemble the entire 565.... I said to myself, there's someone more crazy than I am... 😀
I meant the main caps can be bypassed.
35,000uF +
1000uF +
100uF +
5uF +
0.5uF +
0.1uF
The first (35kuF) and last one (0.1uF) are already part of the stock set-up. You just need the other 4. Make sure they are 100vdc types and orient them correctly...
You need 1 per rail or 8 caps total...
The other ones will go to the rails that feed the OP devices. .. these are not that important as bypassing the main caps.
Also life will go on without these bypass caps, I just mentioned it after I saw you dis-assemble the entire 565.... I said to myself, there's someone more crazy than I am... 😀
- Status
- Not open for further replies.