Heretical 834p mods
The EAR 834p, is a marvellous and clever circuit, in its various optimisations and design features that are often not really understood... I acquired my second one a while ago, my first one was tweaked quite successfully to make very acceptable sound by adjusting the 10pf to 5pf and changing output CF to ECC81, and removing some surplus output potential dividers, and careful valve selection. Lowest noise first, and highest gain second.
However, it went away...
The problem I have seen with these is that they sometimes suffer muddy woolly sound in the bass which tends to spoil the over all effect. The new phono box I recently got suffers this way. Now I am suspicious of feedback loops, and how they work. I suspected that the 834p was maybe running a little short of gain in the lower registers. As I enjoy heresy, and after much consideration, I decided that bootstrapping the second stage ought to give me some more gain for the feedback loop to work. See circuit below for simplified circuit.
So I added a unity gain opamp buffer running from the 30 volt heater (thanks Tim) and capacitively coupled to the junction of an additional 10k resistor placed at the top of the second valve's 330k resistor. This resulted in the mushy muddle becoming worse (pause for a few months). Slightly demoralised, I was pondering an entire new preamp design, and found an interesting circuit using miniature Russian valves, so I decided to simulate, not to bad, then I decided quite easy to tweak the circuit into an 834p, (all be it with 6sl7s gain 70). This gave an interesting result with a reduced shelf at the low frequency end. In simulation this could be fixed by replacing C1 with a 470n. Tried that, and could feel definitely heading in the right direction, all sounding much crisper and tighter, but unfortunately on the big Dovedales, to much low bass. Hmmmm.
On further pondering of the circuit and RIAA correction curve I had a ping moment realisation, that the EQ section was missing the lowest time constant. With my new improved gain circuit, and extra oomph, from the change upto 470n, my bass frequency response was still going up, instead of levelling off at 50Hz!
A quick calculation showed that what I needed in parallel with the 330p was a 10Meg resistor (thanks again Tim). This was quickly fitted, and low and behold I now have a pretty happy sounding setup. My "turntable" at present is a bit mid fi (Sony PS22, definitely better than a 505 tho imo, must do some thing to that boomy lid...), with an old Grado cartridge. I know the setup is working well because I can play almost any of my (dodgy car boot sale) records, of any period and they are listenable, and this is a first in this incarnation of my stereo (this is my third time around with vinyl). The most disconcerting thing is that the music seems almost disconnected from the "record playing mechismo", I am not sure mechismo is a word but its the best comes to mind at the moment.
What I am trying to convey is that traditionally when one puts on a record, one is aware that there is a record playing and it is almost part of the performance, constantly in the background. Whatever that quality is, is largely missing now. For the record, ehem, my previous experiences with this deck were using the Technics SU-9011. The music just sort of is, with surprisingly large dynamic range, of barely audible to irritatingly loud (to other family members) on some classical records, but with no feeling of surface noise or hiss. It is obviously not "realistic". There is still a slightly unclear quality, but is suspect it may be related to the turntable and cartridge. The combination is working pleasingly well. I am very sensitive to miss tracking, and am afraid of listening to classical records because of the horrible grating climaxes that pounce occasionally. I have made it through about 5 classical discs so far without any real show stopping pain, and had my best Joan Armatrading, "show some emotion" performance so far. Even my Glen Miller tribute album sounded half decent, not sure what Ray Conniff was up to though, Cliff Adams is still good, used to sound so lush on Radio 2, before the Top 40 take over of the VHF Network on Sunday afternoons on my Rigonda radio gram......
Never mind. I could go on, but I think you get the picture. Early days and probably more of a beginning than an end. I am contented with this 834p, and I can now enjoy my considerable (inherited) record collection....
Back to the matter in hand. Maybe the designer cunningly engineered the last time constant in some where else, possibly even relying on the second tube running out of gain. I can only speculate at this point in time. But if that was the case maybe it explains the mysterious muddyness that afflicts this circuit from time to time.
Hmm, how about a version with a CCS loaded pentode in the middle position, using the suggested RIAA tweaks, should do away with the need for the "opamp", if the extra gain was carefuly managed... hmm ECC83 and 2 ECF80s..... oh dear, another beginning... brain calm down, need sleep...
However, it went away...
The problem I have seen with these is that they sometimes suffer muddy woolly sound in the bass which tends to spoil the over all effect. The new phono box I recently got suffers this way. Now I am suspicious of feedback loops, and how they work. I suspected that the 834p was maybe running a little short of gain in the lower registers. As I enjoy heresy, and after much consideration, I decided that bootstrapping the second stage ought to give me some more gain for the feedback loop to work. See circuit below for simplified circuit.
So I added a unity gain opamp buffer running from the 30 volt heater (thanks Tim) and capacitively coupled to the junction of an additional 10k resistor placed at the top of the second valve's 330k resistor. This resulted in the mushy muddle becoming worse (pause for a few months). Slightly demoralised, I was pondering an entire new preamp design, and found an interesting circuit using miniature Russian valves, so I decided to simulate, not to bad, then I decided quite easy to tweak the circuit into an 834p, (all be it with 6sl7s gain 70). This gave an interesting result with a reduced shelf at the low frequency end. In simulation this could be fixed by replacing C1 with a 470n. Tried that, and could feel definitely heading in the right direction, all sounding much crisper and tighter, but unfortunately on the big Dovedales, to much low bass. Hmmmm.
On further pondering of the circuit and RIAA correction curve I had a ping moment realisation, that the EQ section was missing the lowest time constant. With my new improved gain circuit, and extra oomph, from the change upto 470n, my bass frequency response was still going up, instead of levelling off at 50Hz!
A quick calculation showed that what I needed in parallel with the 330p was a 10Meg resistor (thanks again Tim). This was quickly fitted, and low and behold I now have a pretty happy sounding setup. My "turntable" at present is a bit mid fi (Sony PS22, definitely better than a 505 tho imo, must do some thing to that boomy lid...), with an old Grado cartridge. I know the setup is working well because I can play almost any of my (dodgy car boot sale) records, of any period and they are listenable, and this is a first in this incarnation of my stereo (this is my third time around with vinyl). The most disconcerting thing is that the music seems almost disconnected from the "record playing mechismo", I am not sure mechismo is a word but its the best comes to mind at the moment.
What I am trying to convey is that traditionally when one puts on a record, one is aware that there is a record playing and it is almost part of the performance, constantly in the background. Whatever that quality is, is largely missing now. For the record, ehem, my previous experiences with this deck were using the Technics SU-9011. The music just sort of is, with surprisingly large dynamic range, of barely audible to irritatingly loud (to other family members) on some classical records, but with no feeling of surface noise or hiss. It is obviously not "realistic". There is still a slightly unclear quality, but is suspect it may be related to the turntable and cartridge. The combination is working pleasingly well. I am very sensitive to miss tracking, and am afraid of listening to classical records because of the horrible grating climaxes that pounce occasionally. I have made it through about 5 classical discs so far without any real show stopping pain, and had my best Joan Armatrading, "show some emotion" performance so far. Even my Glen Miller tribute album sounded half decent, not sure what Ray Conniff was up to though, Cliff Adams is still good, used to sound so lush on Radio 2, before the Top 40 take over of the VHF Network on Sunday afternoons on my Rigonda radio gram......
Never mind. I could go on, but I think you get the picture. Early days and probably more of a beginning than an end. I am contented with this 834p, and I can now enjoy my considerable (inherited) record collection....
Back to the matter in hand. Maybe the designer cunningly engineered the last time constant in some where else, possibly even relying on the second tube running out of gain. I can only speculate at this point in time. But if that was the case maybe it explains the mysterious muddyness that afflicts this circuit from time to time.
Hmm, how about a version with a CCS loaded pentode in the middle position, using the suggested RIAA tweaks, should do away with the need for the "opamp", if the extra gain was carefuly managed... hmm ECC83 and 2 ECF80s..... oh dear, another beginning... brain calm down, need sleep...
Total Comments 0
