I often see a transistor besides VAS transistor (Q107 here). What is the purpose of this transistor. It seems like limiter, but does it contributes something to (good) sounding?
Also, some members are thinking about clipping behavior. What is this "clipping behavior", does it also contributes to good sounding? Does Q107 have to do with "clipping behavior"?
oops.... one more.
Why is the VBE multiplier sits not at the VAS line? The lower VAS CCS drives 1k resistor, not VBE multiplier.
Also, some members are thinking about clipping behavior. What is this "clipping behavior", does it also contributes to good sounding? Does Q107 have to do with "clipping behavior"?
oops.... one more.
Why is the VBE multiplier sits not at the VAS line? The lower VAS CCS drives 1k resistor, not VBE multiplier.
Hi Greg,
These do have a special sound, they stuck out in my mind even though I've serviced most major brands and most of the "classics". I've taken some home too. They look really good. Even my wife likes them.
They don't run too hot. For the design they only use as much current as needed. They sound beter than a Carver Lightstar (had one for years), but the Lighstar is soooo cool!
I have heard many different amps I like, but nothing extremely efficient seems to sound good. I would suggest you take the time at some point in your life to listen to one of these. I think you may be surprised over what you see in the schematic.
Oh, no "Kitsch value" at all. Most people don't know what a Nakamichi is over here.
Cheers,
-Chris
These do have a special sound, they stuck out in my mind even though I've serviced most major brands and most of the "classics". I've taken some home too. They look really good. Even my wife likes them.
They don't run too hot. For the design they only use as much current as needed. They sound beter than a Carver Lightstar (had one for years), but the Lighstar is soooo cool!
I have heard many different amps I like, but nothing extremely efficient seems to sound good. I would suggest you take the time at some point in your life to listen to one of these. I think you may be surprised over what you see in the schematic.
Oh, no "Kitsch value" at all. Most people don't know what a Nakamichi is over here.
Cheers,
-Chris
Gloriously un-subtle design
What a gloriously un-subtle design.
The output stage scares the he-be-gebees out of me.
Those device must be matched at least on three parameters.
IMO that is no way to design a good producable amp.
Also as was pointed out the Miller VAS compensation is er... well plain bad.
I would sugest that any update/clone includes (at a minimum) the following improvements..
1) Better HF limiting - may be an inclusion of a double pole filter that could be easily worked around for those who wantthe orthentic sound
2) Deal with the miller caps on the Sziklai pairs
3) Deal with the loading requirements on the Sziklai drive transistor
4) Deal with the lack of base stoppers on the Sziklai paris
5) PSU noise injection into the input ground due to LTP cascode and bias current reference points
Attached is a PDF of the scans, I thaught it migh be a bit easier to handle.
Brian
What a gloriously un-subtle design.
The output stage scares the he-be-gebees out of me.
Those device must be matched at least on three parameters.
IMO that is no way to design a good producable amp.
Also as was pointed out the Miller VAS compensation is er... well plain bad.
I would sugest that any update/clone includes (at a minimum) the following improvements..
1) Better HF limiting - may be an inclusion of a double pole filter that could be easily worked around for those who wantthe orthentic sound
2) Deal with the miller caps on the Sziklai pairs
3) Deal with the loading requirements on the Sziklai drive transistor
4) Deal with the lack of base stoppers on the Sziklai paris
5) PSU noise injection into the input ground due to LTP cascode and bias current reference points
Attached is a PDF of the scans, I thaught it migh be a bit easier to handle.
Brian
Hi Mark,
In all fairness, the build quality with Nakamichi is higher than some Threshold I've seen. Sorry, but it's true as I serviced both.
There are no reliability problems with the PA series Nak amps. The type II design addressed the over active protection circuits, there was a mod for the original ones available at no charge for a while. I installed many.
As far as what Nelson was and wasn't happy with, I'd rather hear it from him in response. Happy with Nakamichi and happy with these amps are two different things entirely. Actually, I'd love to hear what suggestions he may have. His designs have advanced beyond this series, he's moved on. Hindsight is valuable.
I do not hold the Nak amps up as being the best, but they sound very good. I think in that light they are well worth consideration as a project for those who know what these amps are all about. You could do a whole lot worse.
-Chris
In all fairness, the build quality with Nakamichi is higher than some Threshold I've seen. Sorry, but it's true as I serviced both.
There are no reliability problems with the PA series Nak amps. The type II design addressed the over active protection circuits, there was a mod for the original ones available at no charge for a while. I installed many.
As far as what Nelson was and wasn't happy with, I'd rather hear it from him in response. Happy with Nakamichi and happy with these amps are two different things entirely. Actually, I'd love to hear what suggestions he may have. His designs have advanced beyond this series, he's moved on. Hindsight is valuable.
I do not hold the Nak amps up as being the best, but they sound very good. I think in that light they are well worth consideration as a project for those who know what these amps are all about. You could do a whole lot worse.
-Chris
When I think of STASIS I try NOT to think of the NAKS however
I agree with you in regards to build quality on some of the very older Thresholds. Compares to parts quality used in later Thresholds or to todays parts what you got back then was state of the art for that time and place, you would find similar parts used in Audio Research products of the same era for instance..... but what you ALSO see and get is the difference between a hand built product and an assembly line product. Basically the way a Masaratti is built versus a a Chevy. The prices of used Threshold Amps vs. Nak amps of the same time periods also reflect this very well.... Thresholds always have higher values, sometimes by a wide margin. Although Nelson says that he feels that NAK lived up to the Stasis circuit its still in my vision a *** product through and through. I've also serviced many of those amps but not to do upgrades to the MK-1's but to replace blown outputs and rectifiers and other associated parts, there were not many MK-1's around when I was doing these repairs. I'd have rather sat and repaired Thresholds all day long but they don't have the break down incidents of the NAKS. What I had to repair was I admit pretty common stuff but non the less were problems. I guess if you want the REAL thing you would WANT TO clone one of the originals that Nelson designed not one that some Japaneese engineer thought might pass(no pun intended!) for it.
Mark
Hi Mark,
Okay, your position is rather clear, as is mine. I can not agree with some of your comments are they are not borne out in my experience.
Possibly the lower temperatures in Canada make them more reliable here
. I only ever saw 1 (one) PA-5/5II/7/7II blown in my area (Toronto). In or out of warranty. Very few Thresholds. I feel they are both excellent amplifiers.
I do not think you are doing anyone any service by trashing an amp that doesn't deserve it.
-Chris
Okay, your position is rather clear, as is mine. I can not agree with some of your comments are they are not borne out in my experience.
Possibly the lower temperatures in Canada make them more reliable here
. I only ever saw 1 (one) PA-5/5II/7/7II blown in my area (Toronto). In or out of warranty. Very few Thresholds. I feel they are both excellent amplifiers.I do not think you are doing anyone any service by trashing an amp that doesn't deserve it.
-Chris
Well I was 'off the leash' yesterday! Passionate I guess.
Hi Lumanauw,
Yes Q107 is turned on when the current through the associated resistor approaches 0.6V , and robs the Vas stage of further current. This should occur only at clipping or some other overload.
The Vbe multiplier is moved to the Class A driver (comp EF) so the Vbe's of this stage are not included! The multiplier has a thermistor in the b-c leg to accelerate bias reduction at high temp, possibly overcompensate.
Hope that helps.
Cheers,
Greg
Hi Lumanauw,
Yes Q107 is turned on when the current through the associated resistor approaches 0.6V , and robs the Vas stage of further current. This should occur only at clipping or some other overload.
The Vbe multiplier is moved to the Class A driver (comp EF) so the Vbe's of this stage are not included! The multiplier has a thermistor in the b-c leg to accelerate bias reduction at high temp, possibly overcompensate.
Hope that helps.
Cheers,
Greg
I think the VBE multiplier arrangement like that helps in providing somekind of "Non-TurnOff" mechanism.
The VAS bias through R113 is about 4mA. It gives about 4V at R111 (1K). It gives quite a voltage drop at R116 and R119 (is it R119?) =75ohm, because I think here VBE multiplier will not exceed 2V for this output stage arrangement. That gives about 2V drop at R116+R119.
Consider a positive swing. The VAS transistor Q108 will give current bigger than 4mA. It will make the drop at R111 bigger than 4V. Also this give bigger drop at R116. Consider the VBE multiplier value constant.
This will kind of "pushing" the whole VBE multiplier (Q112) downwards during positive swing, making Q114 (negative half) harder to turn-off, compared to ordinary VBE multiplier (where sits at VAS line directly).
The VAS bias through R113 is about 4mA. It gives about 4V at R111 (1K). It gives quite a voltage drop at R116 and R119 (is it R119?) =75ohm, because I think here VBE multiplier will not exceed 2V for this output stage arrangement. That gives about 2V drop at R116+R119.
Consider a positive swing. The VAS transistor Q108 will give current bigger than 4mA. It will make the drop at R111 bigger than 4V. Also this give bigger drop at R116. Consider the VBE multiplier value constant.
This will kind of "pushing" the whole VBE multiplier (Q112) downwards during positive swing, making Q114 (negative half) harder to turn-off, compared to ordinary VBE multiplier (where sits at VAS line directly).
Mark A. Gulbrandsen said:
or a Lexus, BMW or MB? I would take the later any time of the day (actually I would take the Masaratti and flip it to a sucker and use the money on a couple BMWs).
I don't quite understand the fancination associated with "hand-made". I am sure when it comes to our own money, each one of us would opt for a mass-produced Pentium 4 cpu, rather than its hand-wired discrete version from Italy.
there are lots of things where hand-made has its special value. I doubt electronics is one of those things, tho.
A smaller Nak STATIS
Here is the amp section of a Nak TA-3 receiver. It is STATIS with 75 W RMS per channel, 20-20k, < 0.1% THD, 8 ohm.
TA-3A Amp 120 DPI PNG (1.4 MB)
TA-3A Amp 600 DPI TIFF (29 MB)
Here is the amp section of a Nak TA-3 receiver. It is STATIS with 75 W RMS per channel, 20-20k, < 0.1% THD, 8 ohm.
TA-3A Amp 120 DPI PNG (1.4 MB)
TA-3A Amp 600 DPI TIFF (29 MB)
2. If people want to make a Nak, my view is to let them, I recall on the Krell post people wanting to make something else, and discussing other (potentially better) designs, but they were quickly shut down. hmmmm.... its probably early enough in the thread for this type of discussion but later on.... [/B][/QUOTE]
- However on the Krell thread some options for modification were included. Such as the 2 level bias arrangement.
The owner of the PCB process needs to decide what modifications they want to include.
- IMHO The cloning process also requires a cloning of the PSU. Especially on a design that appears to be sensative to PSU noise.
In a lot of ways this is the most difficult part. That hunk of magnetics is a real bear to clone properly.
Even for volume production maintaining multiple supliers for magnetics is one of the pains in my life.
Brian
- However on the Krell thread some options for modification were included. Such as the 2 level bias arrangement.
The owner of the PCB process needs to decide what modifications they want to include.
- IMHO The cloning process also requires a cloning of the PSU. Especially on a design that appears to be sensative to PSU noise.
In a lot of ways this is the most difficult part. That hunk of magnetics is a real bear to clone properly.
Even for volume production maintaining multiple supliers for magnetics is one of the pains in my life.
Brian
Congratulations bra, you once started one thread about cloning one famous amp and that thread become biggest on whole Diyaudio site!
Be careful when you chose something for cloning…
About this particular Nakamichi amp, I never owned one, but I had a chance to listen and compare it with other hefty amps from that era. Nothing spectacular, but decent sound and excellent drivability. I prefer class a/b efficient bipolar designs, but not this one.
However, I’ll never try to persuade someone for cloning or wiling to build this amp. We are all different and we have different needs and desires.
I remember one thread about cloning Bryston amps. Thread is dead with zilch results. Why? Are Bryston ST or SST amps not worth, or are they inferior compared to NAK or Threshold?
That brink’s me to my last question: Mark, I understand your position to give us advice about something better for us diyers, but after all why by all means must be N. Pass design, or something related to his past companies?
There are a lot of good amps worth cloning, especially for me, because I’m not educated enough to build something all mine.
At my present, I’m enjoying my Harman Kardon HK870 power amp, and HK825 pre, modified and refurbished by me, my On-Semi transistors and other parts for Krell KSA-50 are here so I will have enough fun next weeks. After that I will try to do good Pcb for Bryston 3B SST (all schematics can be downloaded at official Bryston site).
In meantime I will wish you all the best with PA7 clone! Go for it, by all means, any new diy amp is better than nothing.
Chicco
Be careful when you chose something for cloning…
About this particular Nakamichi amp, I never owned one, but I had a chance to listen and compare it with other hefty amps from that era. Nothing spectacular, but decent sound and excellent drivability. I prefer class a/b efficient bipolar designs, but not this one.
However, I’ll never try to persuade someone for cloning or wiling to build this amp. We are all different and we have different needs and desires.
I remember one thread about cloning Bryston amps. Thread is dead with zilch results. Why? Are Bryston ST or SST amps not worth, or are they inferior compared to NAK or Threshold?
That brink’s me to my last question: Mark, I understand your position to give us advice about something better for us diyers, but after all why by all means must be N. Pass design, or something related to his past companies?
There are a lot of good amps worth cloning, especially for me, because I’m not educated enough to build something all mine.
At my present, I’m enjoying my Harman Kardon HK870 power amp, and HK825 pre, modified and refurbished by me, my On-Semi transistors and other parts for Krell KSA-50 are here so I will have enough fun next weeks. After that I will try to do good Pcb for Bryston 3B SST (all schematics can be downloaded at official Bryston site).
In meantime I will wish you all the best with PA7 clone! Go for it, by all means, any new diy amp is better than nothing.
Chicco
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.