I don't see that. According to the schematic in post 1, the cascode bias resistors are connected to the positive rail and earth. There should be no inter-action between the current through them and the input pair or current source.
😕
My bad, you re right, godfrey....
I should clean up my folders more frequently...
Funny thing is that connecting the said resistor to ground
greatly increase THD...
the higher rails for the front-end makes a lot of sense, that is how commercial japanese amps did theirs anyway......
are you recommending commercial Japanese amps?
Most commercial gear aimed at the uninformed masses is utter crap.
I'm confused; does the fact that the commercial Japanese amps used this technique invalidate as crap it or is the technique crap and the fact that Japanese commercial designs used it was a coincidence?
Actually, in the Goldmund it looks like the front-end rails are actually at a lower voltage than the output stage rails. It's inefficient, but does keep the output devices out of their non-linear range.
The 70's amplifier market went for low distortion irrespective of what that did to the performance.
They (the Japanese and others) also went for 3 to 6months obsolescence, and some buyers were sucked into this salesmanship.
It did absolutely nothing to improve nor maintain the reputation of commercial amplifiers, their manufacturers, nor their designers.
What I am saying is that the fact that we see some designers adopting a particular topology says nothing about performance.
They (the Japanese and others) also went for 3 to 6months obsolescence, and some buyers were sucked into this salesmanship.
It did absolutely nothing to improve nor maintain the reputation of commercial amplifiers, their manufacturers, nor their designers.
What I am saying is that the fact that we see some designers adopting a particular topology says nothing about performance.
Hi godfrey,
I think you may have missed the voltage multipliers hidden amongst the decoupling caps.
regards
Yamaha and Sony are just two of those that came up with designs in the early 80's IIRC.......
the information was written up as articles in the defunct AUDIO magazines, while i have not heard those amps, i would not call them crap.......their higher cost probably was the reason it did not get widespread use....
having lower rails in the front-end, the designer was probably contemplating not having the output devices driven to saturation....
hey jaco, i was merely responding to godfrey......😀
i have seen the latest iteration of the design and the front rails are indeed higher......😀
Bigpanda, thanks for the reply regarding the relay.
Omron is now in place on PCB. When it comes to the "extra" leg 4, I drilled an extra hole in the PCB, cut some of the leg and made sure that earthplane does not come into contact with this leg. The extra hole is not in conflict with other print lines on the underside of the print.
By making such an adaptation, those who build this amplifier have three alternatives to the output relay, which gives some more choices for purchasing:
Panasonic: JTN1S-PA-F-DC24V
Omron:G8P-1A4P/1A2P
Omron:G8P-1C4P(leg 4 omitted)
There could certenly be other options.
Now I'm just waiting for the output transistors to appear.
Then the promised testing together with Roar Malmin will take place. Over the next few weeks, I hope.
I've already started thinking about what opportunities there are for improvement with Goldmund clone without changing Alex's print.
Early in this thread, it was pointed out serveral "weaknesses" on both the power supply to the pre stages and the output stage.
Ostripper describes in # 113 a solution that could possibly be an improvement, but there is a voltage multipliers hidden amongst the decoupling caps on Goldmund that confuses me. So, Ostripper, come "on the lane" again and "play" with us. How can we reach the correct voltage and implemet your power supply to the Golmund clone? Do you have a PCB for this power unit??
If such a solution is chosen, will it then be natural to think as follows(??):
1.D/6/7/8/10, C16/19 and R38/39 removed.
2.Positive end is connected to + on C/15 and negative end connected to - on C18.
When it comes to the power supply to the output:Will it be a good idea to increase the condenser capasity?? As it is now, I have 10000uF on both sides. If, what could be a reasonable value??
Yet I have another challenge to those of you who were very critical of the whole design and to Nagy in particular. Now that he no longer appears on this or other pages on diyAudio, it would perhaps be possible to bury the "old battle axes" and get back to some of the "spirit" of this site: Help and inspire each other!!!
I think of disagreements with Nagy about how the grounding of the amplifier should be done. Lee Knatta/ AndrewT had strong opinions and obviously extensive knowledge in this area. Do you ( and others), take the challenge??
Eivind Stillingen
Omron is now in place on PCB. When it comes to the "extra" leg 4, I drilled an extra hole in the PCB, cut some of the leg and made sure that earthplane does not come into contact with this leg. The extra hole is not in conflict with other print lines on the underside of the print.
By making such an adaptation, those who build this amplifier have three alternatives to the output relay, which gives some more choices for purchasing:
Panasonic: JTN1S-PA-F-DC24V
Omron:G8P-1A4P/1A2P
Omron:G8P-1C4P(leg 4 omitted)
There could certenly be other options.
Now I'm just waiting for the output transistors to appear.
Then the promised testing together with Roar Malmin will take place. Over the next few weeks, I hope.
I've already started thinking about what opportunities there are for improvement with Goldmund clone without changing Alex's print.
Early in this thread, it was pointed out serveral "weaknesses" on both the power supply to the pre stages and the output stage.
Ostripper describes in # 113 a solution that could possibly be an improvement, but there is a voltage multipliers hidden amongst the decoupling caps on Goldmund that confuses me. So, Ostripper, come "on the lane" again and "play" with us. How can we reach the correct voltage and implemet your power supply to the Golmund clone? Do you have a PCB for this power unit??
If such a solution is chosen, will it then be natural to think as follows(??):
1.D/6/7/8/10, C16/19 and R38/39 removed.
2.Positive end is connected to + on C/15 and negative end connected to - on C18.
When it comes to the power supply to the output:Will it be a good idea to increase the condenser capasity?? As it is now, I have 10000uF on both sides. If, what could be a reasonable value??
Yet I have another challenge to those of you who were very critical of the whole design and to Nagy in particular. Now that he no longer appears on this or other pages on diyAudio, it would perhaps be possible to bury the "old battle axes" and get back to some of the "spirit" of this site: Help and inspire each other!!!
I think of disagreements with Nagy about how the grounding of the amplifier should be done. Lee Knatta/ AndrewT had strong opinions and obviously extensive knowledge in this area. Do you ( and others), take the challenge??
Eivind Stillingen
IMO the GRoundING problems would be best addressed with some HD Photos of Goldmund boards. Clone to the finest of detail. This is the starting point of the ground Plane.
Krisfr
I agree. That would be super. Bigpanda has given the files that can be used.
Eivind Stillingen
I agree. That would be super. Bigpanda has given the files that can be used.
Eivind Stillingen
I haven't "played" in this thread for a while. In the time of this thread I've built 24 complete amplifiers (latest one in the mongrel thread). I am prototyping 2 MOSFET boards now , am going crazy with Alex's cool PCB software. Grounding issues are "yesterday" with me , split grounding + CLC take care of everything.
Goldmund is all marketing and just minimal engineering. 🙁
On this particular amp , andrew T. has the best idea....http://www.diyaudio.com/forums/solid-state/175289-goldmund-mods-improvements-stability-28.html#post2414035 POST 1375 .... 😀
The only and absolute betterment of this "grand idea" is in my living room , a totally separate and cap multiplied 85-0-85 with a "ground lifted" CT on the small trafo (Absolutely 0 mv ripple for the voltage boards).
OS
No, I did see that but when I did some rough calculations awhile ago, I came to the conclusion that the voltage drops in the smoothing circuitry were enough to reduce the front end rail voltage to below the output stage rail voltage.
I might well have been wrong about that, but it did tie up nicely with two other points:
a) The maximum output voltage swing (according to Goldmund) is significantly lower than the output stage rail voltage.
b) The maximum voltage rating of the specified output devices is not sufficient to withstand rail-to-rail voltage swing at the output.
Unfortunately, I don't think Nagys ever mentioned what the rail voltage is supposed to be. Has anybody actually simmed this amp? Otherwise I guess we'll have to wait till somebody builds it to find out what it does.
That was one of the main issues , the devices were insufficiently derated for reliable operation. It has already been decided "on the other side" (KT's thread) to run the modded version of this amp at reduced rails (60 volts). I still fail to see all this interest in such a cheap , suckarse ,poorly designed circuit. Build a symasym , port it to a lateral OPS ... all done. (with a cascoded IPS , it could run 100v rails ! ) 😀
What will it do ?? It (the goldmund) will "hang" in there for a while ... then fail catastrophically when confronted with a difficult load. How do you think Nagsy makes his $$$$ selling cloned input modules ? .. REPAIRING THEM.
OS
OS
I always wondered about that. Blown output devices I'd expect, but how the #@&$ do you blow up an input stage?
On second thoughts, nevermind - I'm probably happier not knowing.