I just fount the interview with Lohstrom that Jan did. The schematic of the amp in that page looks nothing like the RA 840BX I have. The Rotel looks much closer to the "Basic" amplifier , Figure 3.1 in Cordell except without the Miller comp in the VAS and it has two compensation networks after the VAS. This is the amp Cordell uses as a starting point, where everything gets better. So, after I understand the Hafler, I should switch to understanding this one and "follow along" doing the calculations. It may be easier to understand. It too has features that don't make sense to me, besides a few drawing errors. Like .22 Ohms on each output emitter, but another .22 ohms on the input to the output network. This is the simplest of all my amps, yet the one that "passed". Wonder what that means?
I should get the service manuals for my 951's today and see if there is a family resemblance across 10 years. The 951's at lease have protection circuits in them. Just by parts count, must be pretty simple.
I should get the service manuals for my 951's today and see if there is a family resemblance across 10 years. The 951's at lease have protection circuits in them. Just by parts count, must be pretty simple.
Hi,
Yet in post 1216 you claim the difference is unmeasurable as far as you know (I took exception to the "we"):
I somehow am unable to reconcile these to two statements, they cannot be equally true.
And I am not interested in what you should have said, but in what you claimed, in writing in public.
Measurable in what sense?
For example a blind (not ABX) preference test could be used.
Which one?
In terms of measured distortion between a generic electrolytic exposed to around 2mA AC current (original DH-100 circuit) and my suggestion of around 2uA into a high quality film capacitor?
As a rough guess, a few orders of magnitude.
Between different film cap's, depends on many factors, anywhere between FA and 20dB.
Ciao T
Yet in post 1216 you claim the difference is unmeasurable as far as you know (I took exception to the "we"):
I somehow am unable to reconcile these to two statements, they cannot be equally true.
And I am not interested in what you should have said, but in what you claimed, in writing in public.
Measurable in what sense?
For example a blind (not ABX) preference test could be used.
Which one?
In terms of measured distortion between a generic electrolytic exposed to around 2mA AC current (original DH-100 circuit) and my suggestion of around 2uA into a high quality film capacitor?
As a rough guess, a few orders of magnitude.
Between different film cap's, depends on many factors, anywhere between FA and 20dB.
Ciao T
Hi CopperTop,
Some capacitors do behave in ways that can destroy the accuracy of an amplified signal. As Thorsten has said, the type of capacitor used can make a huge difference in sound quality. Also remember that each capacitor type does have an application, something that the design does very well. Non C0G / NP0 ceramic capacitors would not be a good choice in a signal amplifier where voltage changes across their terminals, but are often the only choice at high frequencies seen in RF work.
How much effect a capacitor has in a circuit will depend. The worst case would be a high impedance point in a circuit that has a large changing voltage across that capacitor. Something like, say the voltage amplification stage perhaps?
Some capacitors are very non-linear when passing a large varying current through them. A good example of this would be the output coupling capacitor used in unipolar amplifier designs (not bridged).
If a sonic difference is so great that it can be heard, for sure you can measure the parameter that is having this effect. You might have to work at discovering which characteristic is at fault, but it can be quantified.
-Chris
Some capacitors do behave in ways that can destroy the accuracy of an amplified signal. As Thorsten has said, the type of capacitor used can make a huge difference in sound quality. Also remember that each capacitor type does have an application, something that the design does very well. Non C0G / NP0 ceramic capacitors would not be a good choice in a signal amplifier where voltage changes across their terminals, but are often the only choice at high frequencies seen in RF work.
How much effect a capacitor has in a circuit will depend. The worst case would be a high impedance point in a circuit that has a large changing voltage across that capacitor. Something like, say the voltage amplification stage perhaps?
Some capacitors are very non-linear when passing a large varying current through them. A good example of this would be the output coupling capacitor used in unipolar amplifier designs (not bridged).If a sonic difference is so great that it can be heard, for sure you can measure the parameter that is having this effect. You might have to work at discovering which characteristic is at fault, but it can be quantified.
-Chris
Last edited:
Hi,
My Service Manual says:
Rotel RA840BX4
That one is definitly an Ottalla/Lohstrom. I would have expected Rotel not to change the whole thing wholesale for a minor revision and that was the schematic I could readily find.
Ciao T
My Service Manual says:
Rotel RA840BX4
That one is definitly an Ottalla/Lohstrom. I would have expected Rotel not to change the whole thing wholesale for a minor revision and that was the schematic I could readily find.
Ciao T
Anatech,
3 out of 4 NAD's I have owned died. No failures of Denon, Hafler, B&K, Apt, Nak, Luxman, Teac, Jvc, Thorens, Kenwood, Onlyo,Sanyo Plus, and I can't remember what else. My NAD C30 CD is doing just fine. Fingers crossed as there are not many plain CD's left that fit on a 12 inch shelf. I did not see corrosion under the glue. No cats in the house though.
NAD's advertising was feeble compared to Curtis Mathis TV's though. "The best darn TV and worth the price". "Quality is Job One" Rule one in advertising, take your worst attribute and advertise it.
3 out of 4 NAD's I have owned died. No failures of Denon, Hafler, B&K, Apt, Nak, Luxman, Teac, Jvc, Thorens, Kenwood, Onlyo,Sanyo Plus, and I can't remember what else. My NAD C30 CD is doing just fine. Fingers crossed as there are not many plain CD's left that fit on a 12 inch shelf. I did not see corrosion under the glue. No cats in the house though.
NAD's advertising was feeble compared to Curtis Mathis TV's though. "The best darn TV and worth the price". "Quality is Job One" Rule one in advertising, take your worst attribute and advertise it.
Hi,
Yes, this is rather different.
Only distinguishing feature, Resistor loaded VAS to reduce open loop gain (the Hafler has it as well...).
Quite brutal.
Ciao T
Yes, this is rather different.
Only distinguishing feature, Resistor loaded VAS to reduce open loop gain (the Hafler has it as well...).
Quite brutal.
Ciao T
I guess I've just been very lucky with NAD, except for the CD player that Corey Greenberg gushed over and which failed early on. I did roll my eyes a bit at the unballasted output devices in the 3020, the ancient 3055/2955's, and as anatech points out the single-ended input stage has some iffy temperature compensation.
The surprise is that Keith Johnson liked it so much, as among the attributes that he prized in those days was equality of positive and negative slew rate, and the circuit is at face value not that symmetrical. We both noted the "gutless wonder" character of the 3020, where "improving" a particular part like making the power transformer larger would probably lead to nasty unintended consequences. And it was sure cheap. Keith had acquired his sample from a friend and got it because he was too busy to repair his own amplifier (as you may know, KOJ was famous for making damn near everything he listened to from scratch --- when I first met him we listened to master tapes played back on his custom machine with his custom tape heads (!) and electronics, through his custom powered speakers. He was running 90V peak signals to the speakers through 300 ohm twinlead. Wacky. Sounded good!).
I hadn't heard that about the adhesive, and I appreciate the information. However I think I will let that sleeping dog alone for the moment, although I mentioned somewhere that I'd bought replacements for the bulk caps when I had had to replace the aux supply ones, the first failure mode absent the inevitable switch contacts wearing out and the pot getting noisy.
As far as the 120 circuit, I did look at the effect of the diode further, and believe/agree that the principal function was to avoid saturation in Q6 (? I can barely make out the schematic) and thus avoid storage time recovery effects. It does look like Q3 gets pretty close to saturation but with no serious effects attendant. I suppose if I am to criticize the circuit I should provide an example of another way to do it But the thing works and made money I presume, so more power to it.
Note added: I just was looking around the room at equipment and recalled that one of NAD's "budget" separates, the "Modular Stereo Preamplifier 1000" had a factory manufacturing defect that led to excessive but not wholly abominable hum, which I didn't have the patience to track down for a long time. I believe it turned out to be a faulty voltage regulator section. I got that piece off the shelf as a demo unit along with one of the stereo power amps, the model 214, about 1997, and they survived 18 months of salt air in Malibu at least.
The surprise is that Keith Johnson liked it so much, as among the attributes that he prized in those days was equality of positive and negative slew rate, and the circuit is at face value not that symmetrical. We both noted the "gutless wonder" character of the 3020, where "improving" a particular part like making the power transformer larger would probably lead to nasty unintended consequences. And it was sure cheap. Keith had acquired his sample from a friend and got it because he was too busy to repair his own amplifier (as you may know, KOJ was famous for making damn near everything he listened to from scratch --- when I first met him we listened to master tapes played back on his custom machine with his custom tape heads (!) and electronics, through his custom powered speakers. He was running 90V peak signals to the speakers through 300 ohm twinlead. Wacky. Sounded good!).
I hadn't heard that about the adhesive, and I appreciate the information. However I think I will let that sleeping dog alone for the moment, although I mentioned somewhere that I'd bought replacements for the bulk caps when I had had to replace the aux supply ones, the first failure mode absent the inevitable switch contacts wearing out and the pot getting noisy.
As far as the 120 circuit, I did look at the effect of the diode further, and believe/agree that the principal function was to avoid saturation in Q6 (? I can barely make out the schematic) and thus avoid storage time recovery effects. It does look like Q3 gets pretty close to saturation but with no serious effects attendant. I suppose if I am to criticize the circuit I should provide an example of another way to do it
But the thing works and made money I presume, so more power to it.Note added: I just was looking around the room at equipment and recalled that one of NAD's "budget" separates, the "Modular Stereo Preamplifier 1000" had a factory manufacturing defect that led to excessive but not wholly abominable hum, which I didn't have the patience to track down for a long time. I believe it turned out to be a faulty voltage regulator section. I got that piece off the shelf as a demo unit along with one of the stereo power amps, the model 214, about 1997, and they survived 18 months of salt air in Malibu at least.
Last edited:
Hello Chris,
Could you define the cutoff, for the early NAD stuff. A friend has 2 mono blocks ( actually stereo ran in bridged mode) driving his ESL63's quads and they work very well indeed , no reliability issues, i don't think they would be consider "early " .
Hi Brad,
No, you weren't lucky. Any money that went in that direction could have been used to buy something that sounded good (for real). However, with the audio press being what it was, thousands of people were lead down that path. It's too bad that misinformation is allowed to be forwarded. NAD didn't make CD players. Those were subcontracted in the far east. Same as the entire line.
Most audio reviewers are people I have little respect for. It's one thing to report on something you actually know and understand, but these people often do not understand even the basics. Just look at some of the combinations that have been recommended over the years! Just pick your own, but my favourite has always been the recommendation that a tube preamp should drive a solid state amplifier "to get rid of the harsh transistor sound". That was a pretty mindless combination. In the Toronto area, coupling various tube amps to horrible sounding Bryston product was popular.
If you see that adhesive, remove it. Remove the parts and jumpers if you have to, but do remove that stuff. It ain't a sleeping dog! If this is the stuff I've been talking about, it doesn't sleep. Like rust, it will slowly corrode surrounding metallic objects as it leaks electrons from one place to another unattended place. Like rust, it never sleeps!
Hi tvrgeek,
BTW, the guy who caused my accident was driving a Ford Escape and apparently not looking out the windshield. Driving is your first job, and a car should never feel comfortable like your favourite chair in the living room.
-Chris
No, you weren't lucky. Any money that went in that direction could have been used to buy something that sounded good (for real). However, with the audio press being what it was, thousands of people were lead down that path. It's too bad that misinformation is allowed to be forwarded. NAD didn't make CD players. Those were subcontracted in the far east. Same as the entire line.
Most audio reviewers are people I have little respect for. It's one thing to report on something you actually know and understand, but these people often do not understand even the basics. Just look at some of the combinations that have been recommended over the years! Just pick your own, but my favourite has always been the recommendation that a tube preamp should drive a solid state amplifier "to get rid of the harsh transistor sound". That was a pretty mindless combination. In the Toronto area, coupling various tube amps to horrible sounding Bryston product was popular.
If you see that adhesive, remove it. Remove the parts and jumpers if you have to, but do remove that stuff. It ain't a sleeping dog! If this is the stuff I've been talking about, it doesn't sleep. Like rust, it will slowly corrode surrounding metallic objects as it leaks electrons from one place to another unattended place. Like rust, it never sleeps!
Hi tvrgeek,
Well, yeah! What scares me is Ford and their driver assist stuff. Ford designs break often, and now they have a car that parks itself? Who's accident is this going to be? If you have been reading, Ford plans to introduce an app (uses existing hardware) that detects unintended lane wanderings and "nudges" the car back into the proper lane. It may also invoke a "stick shaker" like operation to the steering wheel. Mark my words, no good will come for this. The best we can all hope for are software crashes that immobilizes their cars (for the safety of all). They scare me to death.
BTW, the guy who caused my accident was driving a Ford Escape and apparently not looking out the windshield. Driving is your first job, and a car should never feel comfortable like your favourite chair in the living room.
-Chris
Hi a.wayne,
To be honest with you, no I can't. I try to ignore junk.
Every single NAD I have ever popped the top on has been garbage. However, if T0-3 2N3055 / 2N2955 parts were ever discontinued, they might have been forced to use a transistor designed for audio work. Faint hope, the heat sinks they used were designed for the plastic T0-3 case, like the MJL series On Semi parts. They jammed T0-3 parts onto that for years and years.
There are several other brands that are not well regarded that are far superior to NAD products. Even the Sanyo Plus series have better design.
As for your friend, hey. If he's happy, that's cool. You will find isolated cases of all kinds of combinations that people are happy with. The deal is that on average, some brands are better made than others. Some technicians are better than others, same goes for designers, lawyers and anything else you can imagine. Just don't look at one case and use that as proof of a concept or idea. Also, people are very poor witnesses, and they certainly are not instruments where consistency is important. This doesn't mean they are a poor judge of anything, but you can only compare the "goodness" of sound quality with the best you have heard so far, and how familiar you are with a sound characteristic. People will gravitate to what they are familiar with as a natural tendency unless they are aware of a shortcoming and they are dying to part ways with it. People can compare two things well (usually), but can not be relied upon to compare things that are not able to be switched between. Their mindset plays a larger role than you would want to accept as well.
-Chris
To be honest with you, no I can't. I try to ignore junk.
Every single NAD I have ever popped the top on has been garbage. However, if T0-3 2N3055 / 2N2955 parts were ever discontinued, they might have been forced to use a transistor designed for audio work. Faint hope, the heat sinks they used were designed for the plastic T0-3 case, like the MJL series On Semi parts. They jammed T0-3 parts onto that for years and years.
There are several other brands that are not well regarded that are far superior to NAD products. Even the Sanyo Plus series have better design.
As for your friend, hey. If he's happy, that's cool. You will find isolated cases of all kinds of combinations that people are happy with. The deal is that on average, some brands are better made than others. Some technicians are better than others, same goes for designers, lawyers and anything else you can imagine. Just don't look at one case and use that as proof of a concept or idea. Also, people are very poor witnesses, and they certainly are not instruments where consistency is important. This doesn't mean they are a poor judge of anything, but you can only compare the "goodness" of sound quality with the best you have heard so far, and how familiar you are with a sound characteristic. People will gravitate to what they are familiar with as a natural tendency unless they are aware of a shortcoming and they are dying to part ways with it. People can compare two things well (usually), but can not be relied upon to compare things that are not able to be switched between. Their mindset plays a larger role than you would want to accept as well.
-Chris
Chris,
That's why I loved my Morgan. It took everything to drive it. No distractions. Not many parts to fail, but if one did, you stopped as it had no extras.
Better check inside my C30 for goo.
I take it you are not fond of Bryston? They are held in high regard down here, but I have never had one to listen to. People love Macintosh, but I have never been impressed. Probably the best amp I have ever heard, as in I don't think it had any sound of its own was the Aragon followed by the Acrus. Trouble with them is on e-bay, they get almost as much as they were new. Probably a hint. If I could see some schematics, maybe they would help my education. They are probably way more complicated than I am ready for.
That's why I loved my Morgan. It took everything to drive it. No distractions. Not many parts to fail, but if one did, you stopped as it had no extras.
Better check inside my C30 for goo.
I take it you are not fond of Bryston? They are held in high regard down here, but I have never had one to listen to. People love Macintosh, but I have never been impressed. Probably the best amp I have ever heard, as in I don't think it had any sound of its own was the Aragon followed by the Acrus. Trouble with them is on e-bay, they get almost as much as they were new. Probably a hint. If I could see some schematics, maybe they would help my education. They are probably way more complicated than I am ready for.
Hi tvrgeek,
I've always wanted to take a 4B and simply build it right (with a redesigned bias control stage) and see how good I could make it. They suffered along with everyone else who used the complimentary differential pair in their designs. They all failed to carefully match all four transistors with each other. Of course, the is a design that is poorly suited to mass production due to the matching requirements. If you do get a good match there, the design works well. That is where you begin paying a good technician money, the detail work that wasn't done before.
Ebay is another thing again. Don't forget that the selling price represents the maximum price anyone was willing to pay at that time. People who know what is up for bid stopped bidding at a much lower price. Then there is that stuff not working thing. That is my personal favourite.
-Chris
True enough. Did that have a positive ground as well as most other British products? As I recall, they were a good looking car.
Not anything before NRB models. Bryston used a differential complimentary pair, but failed to match the transistors. All the parts they used were the least expensive you could get. They use a place called Electrosonic for parts, and you could easily go through their catalogue at the time and pick out every part - exactly the same part, and they would be the least expensive. That stuff was junk. The face plate would tear away if they were rack mounted and moved, and the bias circuit was defective in it's design. That's why they sounded horrible when first on, then after a half an hour they would be plenty hot just sitting there idling. They sounded kind of okay, sort of by then. They would continue to heat up to the point where you normally could not keep your hand on the heat sink. When visiting their plant on an AES tour, they made it a point they matched heat sink colour carefully. Then they told us how closely matched their parts were, but I saw no test instruments capable of matching anything that closely. When servicing various Bryston amplifiers, it was clear from my own measurements that they were telling a big fat tale. I think they brought out the NRB after Stuart Taylor was hired on. I have not heard anything of their line since though, but the schematics look all growed up now. Looks like real product.
I've always wanted to take a 4B and simply build it right (with a redesigned bias control stage) and see how good I could make it. They suffered along with everyone else who used the complimentary differential pair in their designs. They all failed to carefully match all four transistors with each other. Of course, the is a design that is poorly suited to mass production due to the matching requirements. If you do get a good match there, the design works well. That is where you begin paying a good technician money, the detail work that wasn't done before.
Ebay is another thing again. Don't forget that the selling price represents the maximum price anyone was willing to pay at that time. People who know what is up for bid stopped bidding at a much lower price. Then there is that stuff not working thing. That is my personal favourite.
-Chris
I had a pr of 4B's, early 80's, hated them sonically, switched to Audire, Julius stuff was much better back then ...
Aragon , was good stuff in there price range, actually very good sonics, great bunch of guys too. Mcintosh , only their tube stuff worked for me had a Pr of MC3500's, switched to custom made SS after that and never went back to tubes. Try and get the schematics on any Aragon, best of all the amps you have mentioned todate .
Acurus had decent sonics nothing like the aragon line, my brother is using A150's in his home system, reliable with decent sonics...
The MC3500's were no slouch , special really, last great MAC, IMO ....
Trying to understand the bias in the DH 120. I noticed both channels P1 is set to zero. Maximum bias. Glypt from the factory. Position of P2 suggests the Q9 only carries about 2.5 mA.
Not sure why R18 and R21 are different. Is this based on the difference in dc offset of .2V and the charge rate of the gate capacitance?
Not sure why R18 and R21 are different. Is this based on the difference in dc offset of .2V and the charge rate of the gate capacitance?
- Status
- Not open for further replies.