3u3 cap question
One quick question...maybe two.
Would it be OK to parallel 2 caps to obtain desired value ? I think not but I hope I'm wrong. I have only 2uf caps. So I could use only one of them but is it enough, or make about 4uf by adding another one.
Second question regarding abovementioned:
Could someone point where or explain a little what are dependencies between that cap and the rest of the circuit so next time I don't have to bother anyone
and could make my own decision.
Of course I know it's to block DC but I need more.... what about influence on signal frequency ?
Anybody ?
Thanks
One quick question...maybe two.
Would it be OK to parallel 2 caps to obtain desired value ? I think not but I hope I'm wrong. I have only 2uf caps. So I could use only one of them but is it enough, or make about 4uf by adding another one.
Second question regarding abovementioned:
Could someone point where or explain a little what are dependencies between that cap and the rest of the circuit so next time I don't have to bother anyone
and could make my own decision. Of course I know it's to block DC but I need more.... what about influence on signal frequency ?
Anybody ?
Thanks
Hi,
Assuming you're not asking anyone in particular...
Yes you can, C_tot = C1+C2....Cn
The value of the cap will depend on the frequency you want to pass with respect to the input impedance of the next stage.
When this Zin is unknown go for a lower one than actually common to be on the safe side of things. Safe in the sense of low frequency limit the cap can pass, not security related.
The formulas:
f = 1/(2*PI*R*C)
Where f is the - 3dB point at the low end of the frequency spectrum.
From this we can derive:
C= 1/(2*PI*R*f)
Where C is the value of the cap you want to use.
Naturally if this cap needs to block DC from the previous stage you need to know the DC voltage and the AC voltage swing (which you add) and pick the voltage rating to be safe even if a fault occurs.
Basically, if you want DC to be blocked under all circumstances this amounts to your B+ voltage as there's always a chance that a short occurs.
Note that the latter is often neglected for economic reasons.
Cheers, 😉
Assuming you're not asking anyone in particular...
Yes you can, C_tot = C1+C2....Cn
The value of the cap will depend on the frequency you want to pass with respect to the input impedance of the next stage.
When this Zin is unknown go for a lower one than actually common to be on the safe side of things. Safe in the sense of low frequency limit the cap can pass, not security related.
The formulas:
f = 1/(2*PI*R*C)
Where f is the - 3dB point at the low end of the frequency spectrum.
From this we can derive:
C= 1/(2*PI*R*f)
Where C is the value of the cap you want to use.
Naturally if this cap needs to block DC from the previous stage you need to know the DC voltage and the AC voltage swing (which you add) and pick the voltage rating to be safe even if a fault occurs.
Basically, if you want DC to be blocked under all circumstances this amounts to your B+ voltage as there's always a chance that a short occurs.
Note that the latter is often neglected for economic reasons.
Cheers, 😉
What's the purpose of the 3k9 resistor from log to 6922?
I was just about to order up all the parts for Joe's VBIGC and realized that I don't have the faintest idea what the 3k9 is for.
I was just about to order up all the parts for Joe's VBIGC and realized that I don't have the faintest idea what the 3k9 is for.
Hi,
It is what is commonly called a "gridstopper" (gatestopper in SS land?).
It is there to prevent parasitic oscillations and should have its body as close as possible to the grid input of the tube to be effective.
Cheers, 😉
It is what is commonly called a "gridstopper" (gatestopper in SS land?).
It is there to prevent parasitic oscillations and should have its body as close as possible to the grid input of the tube to be effective.
Cheers, 😉
A few more questions before I jump...
Thanks Fedde,
I'm going to use a 6GM8 instead of the 6922 so that I can the entire sheebang off a 24V battery supply (6.3V heaters) and I don't hear much mention of the need for gridstoppers amongst those that have used the tube in preamps. Ron Welborne uses it for his version of Jack Eliano's Ultrapath linestage without one, so I think I can do without.
What wattage and brand of resistors did you use for the 4k7, 18k, 22R, and 1 meg? 5 watt or 12 watt?
What brand of caps?
I'm thinking of using Caddock mk132 resistors(like Brian GT's kit) and AudioCap/REL Multicap PPMF caps throughout?
The PPMF's have gotten glowing reviews and many think they are just about as good as the AudioCap Theta's at 1/3rd the price.
Still a little on the expensive side, but if the performance is worth it... Should cost me $65 for resistors and and $40 for caps including the 1000uf and 3u3 Panasonic FC's. $100 for the whole amp(not counting batteries or charger)
I hope to try the VIBIGC first with buffer and later try an Ultrapath frontend by just adding an ultrapath cap and 2:1 linestage (600 ohm) opts when I have the extra cash.
Thanks Fedde,
I'm going to use a 6GM8 instead of the 6922 so that I can the entire sheebang off a 24V battery supply (6.3V heaters) and I don't hear much mention of the need for gridstoppers amongst those that have used the tube in preamps. Ron Welborne uses it for his version of Jack Eliano's Ultrapath linestage without one, so I think I can do without.
What wattage and brand of resistors did you use for the 4k7, 18k, 22R, and 1 meg? 5 watt or 12 watt?
What brand of caps?
I'm thinking of using Caddock mk132 resistors(like Brian GT's kit) and AudioCap/REL Multicap PPMF caps throughout?
The PPMF's have gotten glowing reviews and many think they are just about as good as the AudioCap Theta's at 1/3rd the price.
Still a little on the expensive side, but if the performance is worth it... Should cost me $65 for resistors and and $40 for caps including the 1000uf and 3u3 Panasonic FC's. $100 for the whole amp(not counting batteries or charger)
I hope to try the VIBIGC first with buffer and later try an Ultrapath frontend by just adding an ultrapath cap and 2:1 linestage (600 ohm) opts when I have the extra cash.
Hi,
The 6GM8/ECC86 shouldn't tend to oscillate under these operating conditions.
If you have an Oscope it's easy enough to find out anyway.
Personally, if I don't see oscillation or overshoot I don't use any in a preamp.
Joe probably has done measurements so for the 6DJ8 users I'd leave it in place.
As for the rest of your QQ, I really haven't built this circuit neither do I intend to. I'm a tube-a-holic, ya see.
However, the results of Peter Daniels' listening tests on the IGCs he built with various caps and resistor brands just so happen to coincide nicely with what I consider "transparent" myself so maybe you should have a look there.
Personally I think highly of the Multicap RTX series (pricey) and Caddock and Riken Rs.
Ultimately, whether or not you want to try it with expensive passive components or not is of course up to you.
If you want to save on useable battery life you could heat the tube from a dedicated, regulated 6V supply.
I doubt doing so would have a negative impact on the sound, I haven't tried it myself so opinion may vary.
One last word, the battery supply quite often benefits audibly from decoupling with good quality (film) & quantity(++µFs) filmcaps in case you haven't considered doing so already.
Cheers,😉
The 6GM8/ECC86 shouldn't tend to oscillate under these operating conditions.
If you have an Oscope it's easy enough to find out anyway.
Personally, if I don't see oscillation or overshoot I don't use any in a preamp.
Joe probably has done measurements so for the 6DJ8 users I'd leave it in place.
As for the rest of your QQ, I really haven't built this circuit neither do I intend to. I'm a tube-a-holic, ya see.
However, the results of Peter Daniels' listening tests on the IGCs he built with various caps and resistor brands just so happen to coincide nicely with what I consider "transparent" myself so maybe you should have a look there.
Personally I think highly of the Multicap RTX series (pricey) and Caddock and Riken Rs.
Ultimately, whether or not you want to try it with expensive passive components or not is of course up to you.
If you want to save on useable battery life you could heat the tube from a dedicated, regulated 6V supply.
I doubt doing so would have a negative impact on the sound, I haven't tried it myself so opinion may vary.
One last word, the battery supply quite often benefits audibly from decoupling with good quality (film) & quantity(++µFs) filmcaps in case you haven't considered doing so already.
Cheers,😉
fdegrove said:
Exactomundo!
This is probably a bit of a safe value (3K9) as I have no control over individual layout etc. If anyone wants to experiment with a lower value, by all means, but unless experienced in these tricky matters, play it safe.
While I'm at it, I want to thank you Frank, and others too, for your helpfulness towards others on this thread. I assure you guys, I do read all postings. But I go through spurts of highly intensive periods of activity and thus not always able to be as present, or seem present, as I would like. But again, I iterate, I do read it all and grateful for others 'standing in.'
Regards to all,
Joe R.
While you are here, Joe; I have built two versions of your diy valve buffered amp (one integrated with one tx for both amp sections, one with a separate psu and two tx for the amps) and have been extremely impressed by both of them 😉. Both use two separate tx for the valves and filaments.
IMHO the integrated (1 tx) version is slightly 'better' as it gives a softer, warmer sound which I prefer, though some may prefer the fuller sound I got with 2 tx. I just wondered if you would recommend any significant changes in any of the cap/resistor values when using two transformers for the amp sections.
IMHO the integrated (1 tx) version is slightly 'better' as it gives a softer, warmer sound which I prefer, though some may prefer the fuller sound I got with 2 tx. I just wondered if you would recommend any significant changes in any of the cap/resistor values when using two transformers for the amp sections.
Ropie said:
Not sure which 'cap/resistor values' you mean? Are you referring to the 4K7/1n3 combination? In which case I strongly suggest varying the cap value. My position has always been that this is worth tweaking. As you know, it is almost impossible to take into account the number of variations of layout and component quality choices as well as personal taste. For this reason the LPF is oh so useful as it allows you to tune it so suit your system.
Let me know if this is useful and also report back your findings. Get cap values in the range of 390pF and 2n2 - try each one of them and give us a 'blow by blow' account.
Joe R.
Joe Rasmussen said:
I was referring to any of the resistors or capacitors from the entire circuit that may need changing as a result of using two tx for the amp supply, instead of just the the one tx you show in your psu circuit
The LPF cap seems like a good place to start so I will get a couple of different values and try them in both. Currently both my amps use 680pF cheapies so I may try some larger, better quality caps, going up to 2n2 as you suggest. Incidentally, I have read reports of bad things happening to the DC offset if the LPF is left out - why is that?
Hi,
Ropie,
If you plan to use the same type of powerxformer twice as described on Joe's webpages, nothing changes in the component department as far as values go.
Naturally you'll need twice as much PSU components.
I fail to see what the LPF has to do with DC at the output of the chip but maybe Joe or someone sees what I don't.
Cheers,😉
Ropie,
If you plan to use the same type of powerxformer twice as described on Joe's webpages, nothing changes in the component department as far as values go.
Naturally you'll need twice as much PSU components.
I fail to see what the LPF has to do with DC at the output of the chip but maybe Joe or someone sees what I don't.
Cheers,😉
Hi all!
A salute to Joe Rasmussen for sharing his VBIGC in DIY land.
This is my first post in this thread, most of the time I just read. But now I would like to share a friend's VBIGCs that I built and is now being used for his HT setup with excellent results.
Again, thank you Joe, and to all who never get tired of helping others!
Cheers!
JojoD
A salute to Joe Rasmussen for sharing his VBIGC in DIY land.
This is my first post in this thread, most of the time I just read. But now I would like to share a friend's VBIGCs that I built and is now being used for his HT setup with excellent results.
Again, thank you Joe, and to all who never get tired of helping others!
Cheers!
JojoD
fdegrove said:I fail to see what the LPF has to do with DC at the output of the chip but maybe Joe or someone sees what I don't.
Cheers,😉
Hi Frank,
The LPF vs DC offset problem was reported by Nuuk on Decibel Dungeon; I don't know what the reason for it is either. I have ordered a small range of different LPF caps to try out this weekend.
JoJo - great looking work; is one of those amps for driving a subwoofer?
Ropie said:
No, the sub used is a powered sub. I have a bridge parallel OPA549 powering a 12" sub though.
Hi!,
I tested a valve buffered gc as per Joe's schematic. I connected a load, but without any signal. When I turn the pot clockwise, the speaker moves in and when I turn the pot counterclockwise, the speaker moves out.
When I connect an input (cd player), the amp works fine, but when I turn the pot, the same weird thing happens.
Any suggestions would be very helpful. Thanks.
JojoD
I tested a valve buffered gc as per Joe's schematic. I connected a load, but without any signal. When I turn the pot clockwise, the speaker moves in and when I turn the pot counterclockwise, the speaker moves out.
When I connect an input (cd player), the amp works fine, but when I turn the pot, the same weird thing happens.
Any suggestions would be very helpful. Thanks.
JojoD
UPDATE... UPDATE... UPDATE... UPDATE... UPDATE...
These are updated Layout Diagrams for Tube Buffered IGC:
1: When Using A Single Transformer For Both LM3875:
An externally hosted image should be here but it was not working when we last tested it.
2: When Using Separate Transformers For Each LM3875:
An externally hosted image should be here but it was not working when we last tested it.
Comments:
It was found that under certain circumstances there were difficulties in driving capacitive speaker cables (although I don't recommend them anyway). Also some JLTi amps showed signs of instability in what seems like older houses with older wirings. Not exactly sure why this would be so. Analysing National Semiconductors AN-202 Application Notes lead to the above alterations. Now it can drive capacitance much better and also hoping that the other matter has been sorted out. Will find out soon.
These diagrams are drawn 'physical point-to-point' as possible, follow them closely as you can.
Technical Stuff: According to NS, because Opamps do not have a 'ground pin' they gain their ground reference from the power supply, usually but not always from the negative rail. This means that the internal 'comp cap' sees most of this negative voltage. But at HF the slew rate limit as well as lack of feedback can create an error in this part of the circuit, known as the integrator, because it lacks restorative force. This means that that it can start talking inter-stage through the negative power supply because the supplie's pulse response is pour. No wonder it can be sensitive to main reservoir capacitors. If we use good capacitors like Panasonic FC with good HF characteristics, then mount them close to the + & - pins on the 3875 and and tie the ground reference (chassis) and the Speaker (-) return to this same point exactly midpoint as shown in the diagrams above. This cleans up potential HF current path at the negative pin and the load return by keeping the 'signal current loop' short.
Because I use little or no feedback above 200KHz, this makes the potential problem worse, because the closed loop gain is a high 33dB. So this may well be a lot less so with Thorsten's Classic IGC.
Joe R.
JojoD818 said:
Check against the posted diagrams, for errors and also to improve your wiring. You may have a severe stability problem or wiring problem. Also check your filament supply while at it.
Joe R.
- Status
- Not open for further replies.