It might be interesting to see how low level distortion evolves with the idle current in the 2EF class AB amplifier with GNFB. Re resistors are 0R22. Please note that distortion is lower at low level for high idle current (as expected), but rises at higher level above the "optimum" bias scenario. I understand that with some older THD+N analyser this would be below the measurement limit. P.S.: 4ohm load, I never use the highly unrealistic 8ohm load.
Last edited:
Hi Pavel,
Agreed. Yes, I know you were talking about the output stage in isolation.
I have Douglas' books and read the article. It's difficult to keep everything in mind all the time. It sure does help to keep theory in mind while dealing with the practical. Also as Doug pointed out, several other factors come into play with a complete amplifier. Everything is a trade-off, so sometimes the optimal bias setting for an output stage conflicts with other things, or if the entire amp has higher distortion you can back off the bias for lower running temperatures. That gives you higher reliability. There is no point running things hot if the result of that doesn't matter in the whole picture.
I measure at 8R per industry standards, but set bias looking for the notch at 10 KHz, 4R.
Agreed. Yes, I know you were talking about the output stage in isolation.
I have Douglas' books and read the article. It's difficult to keep everything in mind all the time. It sure does help to keep theory in mind while dealing with the practical. Also as Doug pointed out, several other factors come into play with a complete amplifier. Everything is a trade-off, so sometimes the optimal bias setting for an output stage conflicts with other things, or if the entire amp has higher distortion you can back off the bias for lower running temperatures. That gives you higher reliability. There is no point running things hot if the result of that doesn't matter in the whole picture.
I measure at 8R per industry standards, but set bias looking for the notch at 10 KHz, 4R.
Anatech, of course cooler is better, but in your experience and as a general rule of thumb - what would you recommend as max temp on heat sink / transistor casing for long term reliable operation
At 50°C, I wonder. I've seen amps run this warm and they are not long term reliable. Understand the internal case temperature will be much higher and those parts also should run cool. So much depends on the case and environment. There is also a thermal fuse inside the power transformer to think about.
Specifying a temperature would probably bring out all kinds of dissenting opinions. A Pioneer SX-1250 runs at 50 mA per device, and those large heat sinks are warm. I wouldn't run it that hot. Meanwhile, a Bryston 4B cubed runs pretty cool and is one of the nicest amplifiers I have ever heard. It is 300 wpc into 8 ohms. So many variables. Performance is your main goal while being reliable. The 4B cubed is one of the lowest temperature amps I have seen. My Yamaha PC2002 runs pretty cool as well (240 wpc), a very clean amplifier. The bias current isn't that high on it either.
Look at Krell amplifiers. They run very warm, require service often (in the grand scheme of things) but sound nice. I wouldn't own one for those reasons.
Specifying a temperature would probably bring out all kinds of dissenting opinions. A Pioneer SX-1250 runs at 50 mA per device, and those large heat sinks are warm. I wouldn't run it that hot. Meanwhile, a Bryston 4B cubed runs pretty cool and is one of the nicest amplifiers I have ever heard. It is 300 wpc into 8 ohms. So many variables. Performance is your main goal while being reliable. The 4B cubed is one of the lowest temperature amps I have seen. My Yamaha PC2002 runs pretty cool as well (240 wpc), a very clean amplifier. The bias current isn't that high on it either.
Look at Krell amplifiers. They run very warm, require service often (in the grand scheme of things) but sound nice. I wouldn't own one for those reasons.
Would be great if you can upload it to https://www.hifiengine.com/manual_library/parasound/hca-1200.shtml which is missing the Mk I schematic.
I just tried to do this. Unfortunately HiFiEngine does not allow uploads or downloads to non-members. I tried to sign up for an account, but that is disabled. I also tried to write to the admin, but there is no way to contact that person except via the "requests" forum on the site, which is blocked except for those with accounts.
Could I email you a .zip file with the three documents, and ask you to upload them to HiFiEngine, if you have an account there?
I've uploaded the files but it takes some time to manually process them for HiFi Engine.
All the new files from Bruce are named parasound_hca-1200_i_*.pdf where i is for first version of the amplifier.
All the new files from Bruce are named parasound_hca-1200_i_*.pdf where i is for first version of the amplifier.
Last edited:
Thank you very much!!
Everyone: Parasound distributes "Education Packs" of information about their vintage products. See here:
https://parasound.com/blogs/news/tagged/Knowledge
I have sent them my documents, and hopefully those will be included in future releases of the HCA-1200 Education Pack.
PS: here is an additional link
Parasound "Educational Packs"
Last edited:
That seems odd. I can't see why they would put web pages and links up and not populate them. Maybe send a note to the web admins?
I'll try again. I tried both links a few times and was only able to download the 1200.
Same thing. I reported it.
Same thing. I reported it.
Last edited:
Threshold stasis models from the 80's were biased at 100 mA and were reportedly in class A mode for 20% of their rated power.
Nelson Pass has stated that he prefers bias to be set to 50 degrees C (I'm assuming he means his amps only) at the heat sinks.
Nakamichi opted to go with a 40 mA bias on their stasis amps; warm at idle but not very warm. Most likely their decision was based on component longevity.
I'm not knowledgeable enough to know the design differences between the Threshold vs. Nakamichi stasis amps but I'm curious to know if Nakamichi's conservative approach was simply that or were there other design factors to consider.
The Stasis amplifiers run the output transistors outside the feedback loop. Therefore, you need a bunch in parallel running at higher bias to keep the delta in current with music lower to avoid higher distortion. There is a technical reason why the bias current is higher. In addition, Nakamichi used transistors with a more linear beta vs current characteristic that allowed lower bias. Distortion would be still lower at higher bias, but you not only have to keep the heat sinks at reasonable temperature, but as you know, you have to be able to control the bias current. You know how sensitive the bias current is to temperature as far as the bias setting on these amplifiers are.
What Nelson was trying to accomplish is the prevent back emf from the speaker from entering the amplifier via the negative feedback loop. Answer: take feedback from before the outputs, isolating the speaker from the feedback network. Personally, I think it is a non-issue except maybe in extreme cases.
Krell runs much hotter, and internal case temperatures are much higher still. They bake themselves.
What Nelson was trying to accomplish is the prevent back emf from the speaker from entering the amplifier via the negative feedback loop. Answer: take feedback from before the outputs, isolating the speaker from the feedback network. Personally, I think it is a non-issue except maybe in extreme cases.
Krell runs much hotter, and internal case temperatures are much higher still. They bake themselves.
Thanks Chris, I knew that the NAKs used different output transistors than their Threshold equivalents but I had no idea on their effect on bias.
That said, I've read on various websites of NAK owners setting their bias at 80-110 mA to be more Threshold like without negative effects other than considerable more heat.
That said, I've read on various websites of NAK owners setting their bias at 80-110 mA to be more Threshold like without negative effects other than considerable more heat.
Well, they are misinformed, as is commonly the case. I would say the extra heat is a major negative effect. I would never make that trade!