We have Freeview TV which uses the old analogue band 4/5 for a digital TV service. I suspect this is common in other countries? Increasingly I find the radio section very reasonable. The sound is at worse bland whilst other stations rich in a similar way to analogue. The format as far as I know is MP3. What I more suspect is Panasonic have done a first class job of the digital. More interestingly direct from the PVR tuner has a more detailed sound when listening to the recordings in near real time via the hard drive. The data space radio might take is unlikely to need extra compresssion. The sound is fine but surprisingly less dramatic via the hard drive.
I have a state of the art aerial given to me by my brother. Maybe that helps. The Panasonic says 10/10 on signal quality.
Listening to this I come to a conclussion which 15 years ago would have been my first reaction ( now I an more cautious ). The drive systems and not digital is what we hear as errors. I have two Marantz CD players of the same series. CD53 and CD67SE. Using the 20 bit crystal DAC the sound difference between both is remakable. There is a similarity that lasts about 20 seconds. The CD53 isn't worth listening to via the DAC. The CD67SE is excellent and the DAC completes that. How very strange. It is covered by theory. I feel my dislike of CD is how it is stores and reads mostly. Production values are also are very variable. Could be the engineers overly believed the " perfect sound " story. Playing back via a cheap CD player would turn off the their creative tallents. Some Beatles CD's are dreadful. How come they didn't realise. Brian Eno CD's don't sound that way.
Talking to a friend he says TV radio has the bit rates intended for DAB which were lost when more stations were added. It certainly sounds that way. Classic FM who use dreadful compression on FM ( including hum ) seem to use ideal compression on TV radio. It's like they have a Revox in loop. BBC radio 3 still seem to be giving us a weird sound. As people say 75% of the time there is no sound at all. If ever radio 3 had an excusss to compress a little. No one would treat TV radio as a holy cow. Classic FM manage to use recordings from very minor orchestras. They serve the car drivers with light classical music. On FM there is a calculation to match realistic in car dynamic range. On TV radio they seem to switch the compression down. Somehow their low budget sound is not bad at all. When Classic FM was taken to the bank manager he was less than impressed with the business plan. They all became very rich!
In my opinion to try to offer a 70 dB dynamic range via a MP3 format is slightly bonkers. I base that on a microphone only and it's preamp. The BBC seem to aim for that. Their FM feed is better. Perhaps that is still is the service they most want to offer?
I tried to say to a BBC friend who did work at radio 3 all speakers crossing at about 3 kHz will have regretable phase changes in that region. There is no right or wrong way. Some speakers she was trying were to her all wrong. They had a 9.8/10 review somewhere. She felt the tweeter was in the " wrong " phase. I said her training will draw her to the BBC sound. Her LS5/9 to me sound like two drive units for about 10 minutes when listening. This melts away as I get back in the saddle with a BBC speaker. To say they are without obvious defects is to be less than honest. To me KEF LS50 possibly are more correct and cost very much less. The first KEF Uni-Q were not great on sound balance. It is a surprise to hear KEF got it to work.
I built a young friend some near field monitors. They will not sound nice if the signal is less than ideal. Some LS5/9 would be useless to him and last 1 week in his hands. Paul Simon Gracelands sounds stunning on them as does David Bowie. Classical sounds surprisingly good, they never will be used for that. They work in the near field to perfection, they slightly soften at 1 metre. They are like AR18's crossed with Celestion SL6 and near bullet proof. They are very very fast. He has no amplifier under 800 watts so an old Rotel 840 given. That is perfection for them. It has power meters. I have told him red means ear danger.
I have a state of the art aerial given to me by my brother. Maybe that helps. The Panasonic says 10/10 on signal quality.
Listening to this I come to a conclussion which 15 years ago would have been my first reaction ( now I an more cautious ). The drive systems and not digital is what we hear as errors. I have two Marantz CD players of the same series. CD53 and CD67SE. Using the 20 bit crystal DAC the sound difference between both is remakable. There is a similarity that lasts about 20 seconds. The CD53 isn't worth listening to via the DAC. The CD67SE is excellent and the DAC completes that. How very strange. It is covered by theory. I feel my dislike of CD is how it is stores and reads mostly. Production values are also are very variable. Could be the engineers overly believed the " perfect sound " story. Playing back via a cheap CD player would turn off the their creative tallents. Some Beatles CD's are dreadful. How come they didn't realise. Brian Eno CD's don't sound that way.
Talking to a friend he says TV radio has the bit rates intended for DAB which were lost when more stations were added. It certainly sounds that way. Classic FM who use dreadful compression on FM ( including hum ) seem to use ideal compression on TV radio. It's like they have a Revox in loop. BBC radio 3 still seem to be giving us a weird sound. As people say 75% of the time there is no sound at all. If ever radio 3 had an excusss to compress a little. No one would treat TV radio as a holy cow. Classic FM manage to use recordings from very minor orchestras. They serve the car drivers with light classical music. On FM there is a calculation to match realistic in car dynamic range. On TV radio they seem to switch the compression down. Somehow their low budget sound is not bad at all. When Classic FM was taken to the bank manager he was less than impressed with the business plan. They all became very rich!
In my opinion to try to offer a 70 dB dynamic range via a MP3 format is slightly bonkers. I base that on a microphone only and it's preamp. The BBC seem to aim for that. Their FM feed is better. Perhaps that is still is the service they most want to offer?
I tried to say to a BBC friend who did work at radio 3 all speakers crossing at about 3 kHz will have regretable phase changes in that region. There is no right or wrong way. Some speakers she was trying were to her all wrong. They had a 9.8/10 review somewhere. She felt the tweeter was in the " wrong " phase. I said her training will draw her to the BBC sound. Her LS5/9 to me sound like two drive units for about 10 minutes when listening. This melts away as I get back in the saddle with a BBC speaker. To say they are without obvious defects is to be less than honest. To me KEF LS50 possibly are more correct and cost very much less. The first KEF Uni-Q were not great on sound balance. It is a surprise to hear KEF got it to work.
I built a young friend some near field monitors. They will not sound nice if the signal is less than ideal. Some LS5/9 would be useless to him and last 1 week in his hands. Paul Simon Gracelands sounds stunning on them as does David Bowie. Classical sounds surprisingly good, they never will be used for that. They work in the near field to perfection, they slightly soften at 1 metre. They are like AR18's crossed with Celestion SL6 and near bullet proof. They are very very fast. He has no amplifier under 800 watts so an old Rotel 840 given. That is perfection for them. It has power meters. I have told him red means ear danger.
Lets try a new idea and see how dead this co
We so carefullly avoid what we set out to know. Instead we have to endure endless "almost science" which are basically enginerring limits. The difference being that science says what we want and enginneering says what we can have.
I doubt this is a good example as it is only my little pet. EL 34 is my favourite valve even money no object. It looks to be undersized, but isn't. It seems to be underprice and isn't. It even seems reliable. There must be a catch. No there isn't. In fact it has a Swiss army knife of other uses. Maybe the only valve post war designed for task rather than market slot ( 6l6 clones like TT21 etc, bigger and better for radio, not sound ). El 34 proves to me the pentode patent was never beaten ( transisors included if OK with small engineeering differences of feedback and goals ). EL 34 Ironmically best to my ears as a high Mu triode not unlike PX25 although indirectly heated. High gain is a big bonus. A PX4 also if other grid baising used. So similar as to beg a question as to why. Simple, the engineering dictates it. To be reliable one must weld firmly the electrodes. The metal is what it is. It is not a universal enginnering solution to conductivity, thermal stresses and microphoney.
I proposed to JJ she make a TR34 pure triode ( we are not reallly friends and although she likes idea not this one. Silly really as she can not fail to make it work. She and I like to think we know best and that's why it wont be made. I couldn't finish another project she wanted so my way of payback ). She could be lazy and just not fit g2,3. With thought one might have a new g3 in a mid spacing to be like a better 300B if my minds eye curve is correct. Bottoming voltage willl be better when 300 B. As I don't like 300B as an output valve not for me. GU50 has these curves and should be best of breed. Again a TR50 perhaps ? For any who don't know valves almost take the work out of your hands. Transistor are almost as easy I admit Valves almost never give bad results in sensible circuits. Pentode curves are the easiest thing I know of to make optimum. If you have no test gear don't try as you can not know what you did is a sweete spot of universal use or a one off for that device. The sweet spot ususllly is so easy to find and so repeatable even with different valves. Spending a penny to 50 pence willl ensure that if biasing. Anything more usually brakes the design. Some can calculate it. Not really as there are tube limits. Test gear will make a more universal circuit. Valves can be asked to be supurb CCS's ( maybe sink ). People seem not to understand this and convert them to voltage control. Like an FET please don't ignor it's best quality.
Andrew T . If in on this you seem you would be the man to blow this apart. Please do. I suspect if you use valves it was in the past? I am sure like me you trained on them.
Work now so forgive the English etc ( doubtless bits missing when I don't edit ). Take it that enginerring limits extended to make subjective results better. Obvious I guess?
rner is.The lounge is not used much these days! Sound quality seems something which forums seldom risk talking about. Except if welll known themes. We so carefullly avoid what we set out to know. Instead we have to endure endless "almost science" which are basically enginerring limits. The difference being that science says what we want and enginneering says what we can have.
I doubt this is a good example as it is only my little pet. EL 34 is my favourite valve even money no object. It looks to be undersized, but isn't. It seems to be underprice and isn't. It even seems reliable. There must be a catch. No there isn't. In fact it has a Swiss army knife of other uses. Maybe the only valve post war designed for task rather than market slot ( 6l6 clones like TT21 etc, bigger and better for radio, not sound ). El 34 proves to me the pentode patent was never beaten ( transisors included if OK with small engineeering differences of feedback and goals ). EL 34 Ironmically best to my ears as a high Mu triode not unlike PX25 although indirectly heated. High gain is a big bonus. A PX4 also if other grid baising used. So similar as to beg a question as to why. Simple, the engineering dictates it. To be reliable one must weld firmly the electrodes. The metal is what it is. It is not a universal enginnering solution to conductivity, thermal stresses and microphoney.
I proposed to JJ she make a TR34 pure triode ( we are not reallly friends and although she likes idea not this one. Silly really as she can not fail to make it work. She and I like to think we know best and that's why it wont be made. I couldn't finish another project she wanted so my way of payback ). She could be lazy and just not fit g2,3. With thought one might have a new g3 in a mid spacing to be like a better 300B if my minds eye curve is correct. Bottoming voltage willl be better when 300 B. As I don't like 300B as an output valve not for me. GU50 has these curves and should be best of breed. Again a TR50 perhaps ? For any who don't know valves almost take the work out of your hands. Transistor are almost as easy I admit Valves almost never give bad results in sensible circuits. Pentode curves are the easiest thing I know of to make optimum. If you have no test gear don't try as you can not know what you did is a sweete spot of universal use or a one off for that device. The sweet spot ususllly is so easy to find and so repeatable even with different valves. Spending a penny to 50 pence willl ensure that if biasing. Anything more usually brakes the design. Some can calculate it. Not really as there are tube limits. Test gear will make a more universal circuit. Valves can be asked to be supurb CCS's ( maybe sink ). People seem not to understand this and convert them to voltage control. Like an FET please don't ignor it's best quality.
Andrew T . If in on this you seem you would be the man to blow this apart. Please do. I suspect if you use valves it was in the past? I am sure like me you trained on them.
Work now so forgive the English etc ( doubtless bits missing when I don't edit ). Take it that enginerring limits extended to make subjective results better. Obvious I guess?
Last edited:
Which may be why people don't like it so much.
Looking at how most valve amplifiers are run without their safety covers for show, KT88s and 300Bs have size for the bling factor
That's a good point. With valves more so I feel if using a generalistaion of audio. They are perfect in what the internal space alllows. In the linearity graphs it seems ones sees this spacing as clear evidence. Which version of not quite linear is personal. The speed, air and layering seem to overcome any smalll linearity questions. sometimes one hears a valve with no transformer coupling. It can take your breath away. The Croft could and it was cheap enough.
I hear Rega have a £5000 MC PU. It seems they must have a big turntable coming?
I hear Rega have a £5000 MC PU. It seems they must have a big turntable coming?
I think that must be fashion and unsuited circuits. Many designers love them when not sellling. Are we snobs. Marantz Model 9 willl eat nearly anything. On a blind test I think it would be bought. It is a Willlaimson more or less with not KT valves. I strongly dislike the KT88 and can measure why. In an optimum circutit without feedback the thrird harmonic is raised. RL 34 is double THD but with perfetct spectrum. My kind of sound. A cheap pyrex tube not only loooks good but can help sound.
I might have graphs. I had to clear my cookies for goodness knows what reason so graphs might take time.
See 300B verses the G2 EL 34 can work quite welll if gain OK. No KT88 of note. KT88 and I are exactly the same age almost to the day. 60 soon. I'm told vast numbers of KT88 were dumped in the Thames. When first dumped most were very OK.Flashover I would guess as the reason.I would imagine no bosss knew this. It's suggested it was done for fun. If so there may be perfects.
I do find a pentode driver for G3 triode works well. The RH34 if used with EF184 is interesting as you will get 3 free lunches of which one is gain. The RH has shut feedback to make the least know triode.
Hopefully some new ideas tomorrow. I work all my life with transistors so they are not of no interest.
Many EL34 amps are very linear. Some use no obvious feedback. I think the person showing this has a passion for G2 connection. Publish curve G3 triode curves for EL 34 are unlike this. What I think we see is the elements as they would be before triode wrapping. Seems a bit speculative as it is not a practical use. No real idea as alas the notes are lost.
The 300 B curves are very nice.
The 300 B curves are very nice.
I hope Alex Kitic is OK about me promoting him here. He was excluded from DIY Audio.
The design resembles a RCA circuit for 807 in 1938 ( Fender ? ). 807 is not exactly like 6L6 or 6L6GC when tested which is bizare if data is correct. They were supposed to be the same. Doubtless the anode cap is the reason and cooler running anode giving power and voltage headroom. Doubtless also the engineers saw the need for an RF valve to be special. That said in most circuits 807 would be indentical. A radio AF amp for example. RCA gave out a version of the Kitic amp in the 1960's It was more or less the same circuit as the Dynaco with this added Rp feedback. Dynaco I fear should have used the better circuit ( Perhaps like me listening counts and they rejected it). Dynaco actual circuit is the RCA publicity handout for improved TV valve 7199. Hard to say if RCA and Hafler worked together on it. 7119 has an audio screen. The pentode is high Mu. It's triode DC coupled as unity gain phase splitter. Of all the valves not to be cloned 7199 is one I don't understand. JJ could.
The RH34 feedback circiut mistakenly is said to be anode to anode ( R7 ) by many, it is able at great speed to iron out nonlinearity. The very high RP ( Anode plate reistance ) of ECC 81 makes this possible. At a guess a 3 : 1 ratio valve to shunt resistor, A cathode capacitor to V1 is a no no for high Rp as it is a bootstrap shunt. EF184 TV pentode looks like an ECC81 when triode. If the pentode curve is usd the overal results look like 300B for the pair. If a ECC82 is used the distortion is OK, it could be better than ECC81. Not vastly better as people insist. This is because the I to V function is not listed in books and the circuit looks 90% coventional . The ECC82 gain is less. A free lunch is lost. It seems ECC81 is OK against all commers if wanting life simple. I have to say some of my biggest failiures were all triode SE amps. On the otherhand a SE in PP out fantastic, it only worked in all triode. Pentode in, triode out is fine in SE.
The RH34 is a transistor like output pentode with local feedback ( Like VAS almost ). As with the transistor a very OK triode like output is had for very few components. An I to V converter if looked at the right way. The shunt is V to I. All proved by converting to less than transconductance when ECC 82. My favourite version I built was 2 x ECC 82 with 82% UL Feedback to G2. Loop Feedback via output coil and ECC 81 cathode ( 18 dB ) . From memory 0.6 V in 8 watts out. 0.1 % THD with nice harmonics. The loop was the same on or off except 1% THD more or less. It shows simple loop feedback can work very well. 18 dB is plenty of feedback. Kitic would't want loop feedback and nor do I need it. My idea was to equall the Pye Mozart with off the shelf parts. I think I beat it. I have it somewhere if asked.
This Kitic amp to me looks to have reduced the Rp feedback considerably ( too lazy to calculate it, bet I'm upside-down in my ratios ) . The 807 being easier to calcuate in RH807. Alex was banned because he was so unpleasent if people changed the design. This is odd because after the ban he follows the route other were saying. I think he thought people were saying he was stupid and a copy cat. I always say where a design comes from. I don't design things for days when a ready made circuit is a building block. The no no ECC83 also used. Nothing worse than a hypocrit. All the same best of luck to him. ECC 82 cascode is as good as ECC 81 any day. Alas two bottles used.
Even the grid leak is about usual . No wonder people thought he was wrong. I built it and it did work. The feedback seemed about 10 dB. Kitic claimed V2 Rp as 900 R. This if true is interesting as triode shunt is 1200 R. If so antiphasing is at work. The Rp was for pentode in shunt ( KT88 ? ). I had about 5 % THD in his cicruits supposed to be 2%. He seemed uninterested in that. 1% should be possible. Perhaps V2 Rp was his real aim. Very wise if so. This simulation looks right on this one.
I have just had the real flu. Doubtless I could have writtten this better.
Last edited:
I think the Kitic dsign is known for it and Alex's preference as it gives a bit of boost. The fo of the transformer is dominant. No loop feedback. Rp V2 is said to be 900R. The load 8R . That gives 10.60 R Z out. I suspect damping factor off 2.3 which just about can control the cone if 1955 Wireless World's real example amplifier is to be believed. Itis the Kitic sound.
RH Amplifiers: RH-307A
RH Amplifiers: RH-307A
I like Schade is implemented there, but so far i´ve seen it done with pentode drivers.
I hope Alex Kitic is OK about me promoting him here. He was excluded from DIY Audio.
Makes me worried a bit, because pentode driver has high ra; it doesn´t interfere much.
Ecc82 is kinda opposite.. Will Schade fb work by same settings.?
about Citation XX
Citation XX is 99% as philosophy like citation 22 ,XII ,also in 7100,990 and others you can see the same engineer thought.The think that make xx so excellent is the assebling (temperature distorsion ,high currenr source ,RF ecc).Also on the power PCB you see particular attention to bias( 2 relays,8 transistors ,in direct contact with immense heatsink).Lets try to find the driver and power schematic from some service center ,especially in USA or Japan. Harman kardon has made 1000-1500 only pieces and MILLER ,OTALA(is death?), and a japanese named Shirasura ,they dont want to publicate there creation!!!.We can make a clone ,because find that amplifier is impossible think.A photo from near distance is enough. Dont remember 30 year evolution in semiconductors can make a better unit...Also a revolution amplifier for his semplicity and design is accoustat ,all mosfet tnt 200 or 120.
- Status
- Not open for further replies.