| ashok |
I've been looking at many tubed circuits for an MC cartridge.
The 6DJ8/6922 seems to be used by some.
I was wondering if the 6922 can be used without any cathode resistor. Only one resistor from anode to supply ( say 68K ) and one resistor from grid to ground ( say 2 K ).
It seems to work in a simulation but is this a meaningful configuration for an MC input ( max input say 20mV on severe peaks )? I remember seeing some tubes with no cathode resistor somewhere on the web. A zero volt grid bias works for some situations in some tubes. Since MC signals are very low and I assumed they may work in the linear region for low negative swings at the grid , typically less than 2mV max for a cartridge sensitivity of 0.2mV @5cm/sec .
Any opinions?
Cheers. |
|
|
| fdegrove |
Hi,
No problem, it works fine and the ECC88/6DJ8 family is relatively linear with Vg=0.
Please note that all tubes need to be selected for low noise.
Here's one that I use on a daily basis:
Cheers,;) |
|
|
| marijan |
| Use a searching engine - tipe 6922 or mc-hamer if I recal corectly. You will find two almost same designs of mc step up signal circuit. Hampton mc using 84 reg. DC voltage, or another by Frank using ccs PS. Grid to ground should be 10-150 ohm (cartridge ?), anode res. depending on PSU typ (paralel 2x 47kOhm in Hampton). No cathode resistors on boath. |
|
|
| ashok |
Hi Frank and Marijan,
Thanks for your really fast response. I will try out the circuit. I have to get a pcb made first ( at home ).
I might be able to select the tubes as I have a few at hand. The CCS may be a problem though. I have some other devices. I will try those . But again I might prefer to try a plain resistor.
I remember Hugh saying that he prefered the sound of a resistor load to a CCS load. Am I right ?
Thanks again.
Cheers.
Ashok. |
|
|
| fdegrove |
Hi,
| quote: | | I remember Hugh saying that he prefered the sound of a resistor load to a CCS load. Am I right ? |
Perhaps he said so in general, but here it's essential to have very high PSRR.
Any CCD or JFET having an Idss of 1mA should do as well.
The BF244As I select as well for equal current and low noise.
I never use any PCBs as I feel the can degrade the sound.
Moreover, by using P2P it leaves me room to fool around with the coupling cap and loadresistor for instance.
Marjan built the Hampton MC which should be a fine cct for all but the lowest output MCs.
Cheers,;) |
|
|
| AKSA |
Yes, Ashok,
I did say that! Different sonics......
I've tried both and much prefer resistive loading.
A tube is effectively a variable resistor, a rheostat, with its resistance controlled by a high impedance voltage at the grid.
If the tube operates in constant current (as it does, within narrow confines, with an inductive load as well), then the voltage output at the plate is extremely linear because the control voltage is not changing the current. It's plate voltage output at constant current for different grid voltages after all. This makes for great accuracy, and complete transparency, something many love. However, with a plate resistor, generally fixed between two and four times the plate resistance, both the plate voltage and the current are changing with signal. This ain't so linear... :eek:
This means that there is more distortion with a plate resistor. Oddly enough, since the introduced distortion is H2 and H3, both commonly found in music, this richens the sound and many people love the sound, muttering things like, 'Ah, there's that tube sound, don't you love the rich structure of the music?'
For me, it's like wine. People talk red but drink white. They hate distortion, go for vanishingly low figures, but LOVE tubes. And when you remove the hood, you find a string of plate loaded triodes. To me, this is the clincher.
So, my point would be this: why go for a CCS with a tube so you finish with complete transparency, like excellent SS?? I say the hamburger is a washout unless it has egg, onions, hot chilli sauce and a dash of mustard...
Nice circuit, Frank! :devilr:
Cheers,
Hugh |
|
|
| fdegrove |
Hi,
| quote: | | Nice circuit, Frank! |
Thank you very much, Hugh.
Odd as it may seem, in the context of my all tubed gear, this CCSed part sounds the most tubey of all when heard in isolation.
Something I don't quite understand but seems to negate the tendency of most MCs towards a rising top end nicely.
Why it sounds like this, especially with a tube that can veer towards stridency like some ECC88s can is beyond me though.
Thanks to the CCS it also gets away with the seemingly rather poor PS whereas the phono stage proper is tightly regulated and has tremendous dynamic range .
So much so that it puts any CDP based system to shame.
Mind you, this circuit is obviously sensitive to layout which is one more reason I'd recmmend P2P here.
Certainly the Hampton is a candidate for improvement too as I think it was originally designed as a mike preamp service?
Again, thank you for the compliment.
I just hope many vinyl addicts will build and enjoy it.
Cheers,;) |
|
|
| fdegrove |
Hi,
| quote: | | So, my point would be this: why go for a CCS with a tube so you finish with complete transparency, like excellent SS?? I say the hamburger is a washout unless it has egg, onions, hot chilli sauce and a dash of mustard... |
Fair enough, an approximation of transparency can be achieved by marrying tubes and SS...in a way but what about dynamic range?
If I were to design SS devices I'd opt for high voltage device as they seem to sound more correct to me.
With tubes I'd go the same road and I'm talking 300VDC+ voltages, the diffrence is often made by the PS.
Cheers,;) |
|
|
| SY |
What kind of voltage do you normally see on the plates? What's the actual gain when loaded down by the ubiquitous 47K input resistor in the following MM stage?
Parallel tubes are the only way to go in this type of topology. |
|
|
| fdegrove |
Hi,
| quote: | What kind of voltage do you normally see on the plates?
|
In my design, if you can call it mine, 24 to 40 VDC can be had at Vg0 for the ECC88/6DJ8.
| quote: | | What's the actual gain when loaded down by the ubiquitous 47K input resistor in the following MM stage? |
Around and about 20dB depending.Why?
You don't need to care about that 47K resistor anyway...
| quote: | | Parallel tubes are the only way to go in this type of topology. |
Yes, which is why I insist on matching up to the moon, not just the tubes but the CCSs as well.
There's little point otherwise....
Cheers,;) |
|
|
| SY |
From the schematic, it looks like there will be 22V on the plate, maybe a skootch less.
Understanding gain structure in a system is essential. Different cartridges have different outputs; different MM amps have different gains. So, gain is a pretty vital piece of info. |
|
|
| ashok |
Hi Frank, Hugh and Sy,
I simulated the amp with a 6DJ8 .
1. The circuit had a 2k resistor from grid to ground and a 39 K resistor from anode to supply of 200 V. Load was 47K with 100pF in parallel ( my MM input stage impedance) coupled by a 1 uF capacitor.
2. The second circuit had a current source using an IRF9640 at the anode. Everything else the same.
The sim showed very low distortion but the CCS ckt had a slight increase in distortion as the harmonics went above the 4th and then started to drop again after the 7th or so.
The resistor load on the other hand kept dropping as the harmonics went up.
Wouldn't this result in a smoother relaxed sound from the resistor loaded circuit ? Or can the sim not be trusted ?
Thanks. |
|
|
| marjan |
Hampton
pay atention on the grid resistor values. |
|
|
| AKSA |
Ashok,
Without deprecating Frank's circuit, I believe you are right. Highish levels of H2 and H3, rapidly dropping thereafter, certainly do give a relaxed, full bodied sound, and are the hallmark of triode circuits.
If you listen to a lead guitar without amplification, you are struck by the thinness of the sound from the string. However, pass it through a Fender, with VERY high loading of the plate, and the sound becomes something very different. Many audiophiles like this sound, and indeed it is routinely associated with electric guitar.
The higher harmonics, albeit very low in magnitude, are discerned over a listening session as an almost subliminal unease. They reflect the highly non-linear behavior of current sources; we know that the triode is essentially blameless in the production of higher harmonics, at least past the 5th, and so a reasonable deduction is that this increase is in fact produced by the mosfet. It is not widely known but current sources have huge impact on sonics in routine amplifiers, and care must be taken that the CCS does NOT contribute to objectionable distortion. Any noise appearing on the base/gate of a CCS device will be amplified, as the device is in common emitter/source, so special care is needed to scotch this insidious, broadband noise. (Ashok, you are also seeing broadband noise, not just distortion products).
I feel jfet current sources are not the best, but they are the simplest. Fred Dieckmann has some strong views on CCS, and might be roped into this discussion......
You there Fred?
Cheers,
Hugh |
|
|
| ashok |
Hi Hugh,
It would be nice if we can discuss this thoroughly .
I did notice that tubes have the least possible distortion with very high anode loads. The resistance is probably closest to being linear though not the highest load. CCS on the other hand can give very high ac loads but as you say , will also contribute their nonlinearities and other characteristics. So an otherwise potentially lower distortion will be swamped by their inherent weaknesses.
What would be interesting is to see how audible this is going to be. Listening tests can be very long drawn out affairs !
I can build the two circuits in the meantime and see if I can hear any differences easily. The resistor load is attractive as it is very simple and can't have any non linear or hf related problems.
I plan to use an Audio Technica AT30 MC cartridge that I currently have. Its a bit old but works well. I also have an AT step up transformer . Will have to do lots of listening I guess !
Thanks,
Ashok. |
|
|
| fdegrove |
Hi,
| quote: | | From the schematic, it looks like there will be 22V on the plate, maybe a skootch less. |
I measure 23 to 24VDC after the CCS, the B+ is 24VDC.
| quote: | | Understanding gain structure in a system is essential. Different cartridges have different outputs; different MM amps have different gains. So, gain is a pretty vital piece of info. |
Gainscaling certainly is....I usually remove the 47K gridleak resistor of the MM stage.
By doing so the 1M bleeder at the output of the headamp becomes the gridleak of the phono preamp.
Soundwise I prefer it that way. I remember hooking up a tuner once to the MC headamp, taking the signal straight to the amp and it sounded rather too sweet to me.
In short, with the CCS it certainly doesn't sound like your typical 6DJ8 at all, no strident ear piercing.
Cheers,;) |
|
|
| jewilson |
Frank,
What differences have you found running the heater on DC vs AC. I was wondering which had the lowest noise, distortion and do they sound different.
thanks |
|
|
| analog_sa |
| Not Frank but i certainly have an opinion on this :) You can't really use AC at these levels. It does sound different of course and even if you only use DC you'll notice significant difference in sound depending on use or not of regulator, type of regulator and type of capacitors. Apparently the cathode-heater capacitance is sufficient to couple all sorts of garbage. |
|
|
| jewilson |
I have heard that AC sound's warmer than DC, however that might just be some one's BS. It does make since that a low noise dc regulator might be best on the gain stages, however I don't really know.
I have a couple of 50 watt Tube amps that I'm upgrading. I though that maybe changing to DC regulation for the input tubes and phase converter would be an improvment. |
|
|
| fdegrove |
Hi,
| quote: | | Apparently the cathode-heater capacitance is sufficient to couple all sorts of garbage. |
Indeed it is...
The easiest way to find out is to simply tap the signal off the DC heater circuit and listen to it.
You won't like the sound of it but it proves that the coupling is for real.
When it works one way, it also works the other way around as it is capacitive coupling.
So, you'd want that heater supply as clean as a whistle especially at this kind of levels.
| quote: | | I have heard that AC sound's warmer than DC, however that might just be some one's BS. |
It does sound different, especially DHT or DHP will even show a different set of curves when you trace them fed from either AC or DC.
In general, I'd like to mix both ways of heating in a system.
Most of the time you just don't have a choice, unless you want to run the heaters off battery supplies, of course, which is a rather clean DC voltage depending on the type of battery you use.
AC is better for the tube's heater as it has a cleaning effect preventing notching. Whether you like AC or DC heating better is a matter of taste but if good signal to noise ratio is important then I'd recommend the use of DC and as Peter says different regulators have different sonic signatures.
Another way to go about is to use both a voltage reg. and a current reg.
It is said to sound the same as AC heating on a DHT without the induced hum penalty.
Personally I haven't tried that yet on my 2A3 amps so it's just hearsay as far as I'm concerned.
| quote: | | I have a couple of 50 watt Tube amps that I'm upgrading. I though that maybe changing to DC regulation for the input tubes and phase converter would be an improvment. |
For an amp that's not humming I'd put my money elsewhere first...
Cheers,;) |
|
|
| duderduderini |
Hi Frank
I built the circuit you have shown in post #2 of this thread. When the outputs of the ccs are unloaded (ie not connected to the 6dj8's) I get a reading close to 24V dc but when the 6dj8's are connected the voltage is down to 12.1V.. Is this ok? The trannys are 1 amp devices so i figure its not that. All components are as specified in the circuit diagram... Sounds good either way though
Thanks for any help
Nick |
|
|
| fdegrove |
Hi,
I assume you measured this voltage between plate and ground?
It's nothing to worry about for as long as it's consistent between channels.
I suspect the BF244As you've used are what's causing it in conjunction with the load.
These diodes vary quite a bit from one manufacturer to another, some only causing a voltage drop of a few volts, some dropping about half the rail voltage.
Different tubes will also make this voltage vary which is to be expected.
IME, the ones you have now are doing a good job and I'm sure the sound is as it's supposed to be.
Enjoy, ;) |
|
|
| duderduderini |
Hi
Thanks for that
It does vary by using different tubes
I got the 244a's from mouser.. i think they are fairchild.. will they vary from one to another or is there a particular brand i should get ( i have about 20 of the fairchild ones)
Thanks again for your help
Nick |
|
|
| fdegrove |
Hi,
There's isn't a particular brand I favour but since the BF 244A is obsolete for quite some time now it's not uncommon to find rebranded other stuff that masquerades as a BF 244A.
They do tend to vary a little from one another as well however and if the difference is too great I'd suggest replacing it with one that's within range.
In general terms, the greater the voltage drop the better they work and the better the circuit will sound.
The BF244A is quite sensitive so don't go swapping out tubes with the headamp turned on, allow it to discharge properly first.
Cheers, ;) |
|
|
| duderduderini |
Hi Again
So what if i did swap out tubes whilst they are on (theoretically of course)
Should i change them?
I promise i wont do it again
Nick |
|
|
| fdegrove |
Hi,
If you were to inadvertantly swap tubes with the circuit on you should check the voltage across plates and ground again.
If it's too high, say 20V or higher then you'll know the CCS is toast.....
Ciao, ;) |
|
|
| duderduderini |
Hi
I checked and its still 12V but thanks for the advice.
Whats your opinion on the amount of filtration on the circuit? I am tempted to run the 24v section off 2 12v batteries in series.. at least with the heaters run off mains power the drain shouldn't be too much.
Regards
Nick |
|
|
| duderduderini |
Hi Again Frank
I have the mc head amp described in post #2 feeding into Jim haggermans Comet octal phono stage and then the line stage is the srpp 6sn7 cicuit described here:
http://www.tubebuilder.com/images/s...ps/6sn7_sig.gif
For reason of copy right i sha't include the circuits.
In order to get the match between the headamp and the comet octal phono, the resistor to ground out of the headamp is 47k as is the resistor to ground on the input of the phono stage of the haggerman. I tried 1M but rather than mains hum it was like a bassy hum, almost a vibration... likewise in the hagerman phono stage which has a 300k resistor to ground on its output, i found less of this noise on using 150k..The output from the phono preamp(haggerman) goes to a 4 way selector switch into the 6sn7 srpp line stage.
Being new to the tube scene i still dont get input and output impedances in the sense that lets say on the input of the comet octal phono stage we see a 47k resistor to ground.. so is that the input impedance? Likewise the 6sn7 has a volume pot directly connected to the grid so is the input impedance calculated by measuring the r values to ground of the pot OR is it all tube impedance?
What sets the output impedance or how do we match from one stage to another?
This to the experienced tubers of this forum will probably be a dumb question.
How does one match all of the stages?
Thanks in advance for your time
Regards
Nick |
|
|
| fdegrove |
Hi,
The 1M ground resistor is there to discharge the circuit when switched off. It does not in any way determine the Zout.
If you replace it with too low a value it will influence the input impedance of the next stage. For example: 47K of your phono input + 47K Rout of the Headamp will make an effective input resistance of 47K/2 = 23K5. This low value may alter the behaviour of your phono stage.
IOW I'd leave the 1M output resistor as is.
I'm not familiar with the equipment you're using other than by name but if you want to check the headamp for hum related problems, short the MC inputs and see if there's noise on the outputs.
This is a very sensitive circuit after all so any hickups in the circuit layout may be a cause for hum or worse, RF pickup.
Putting a grounded metal shield between the xformers and the tubes may reduce things to inaudability.
Using shields on the tubes proper may help too.
Bottomline is, layout is very critical for best results.
As for battery PSU, I've tried this but it didn't make any difference at all. PSRR of the circuit is extremely good even though it may look simple at first glance.
Ciao, ;) |
|
|
| duderduderini |
Hi Again frank
Ok will do re 1M.
I have a seperate outboard psu for the headamp. I have tube sheilds and am thinking of placing the head amp unit in a copper enclosure.. How does one discern between rf and other nasties?
Might I say its a dam fine sounding headamp.. really detailed and warm.. beat the daylights out of the transformer set up i had.
Thanks |
|
|
| duderduderini |
Hi Frank.
I would like to say thanks so much for your help and above all for designing this head amp... its great. I can now enjoy my vinyl again.
warm regards
Nick |
|
|
| fdegrove |
Hi,
Glad you enjoy it.:cool:
Cheers, ;) |
|
|
| duderduderini |
Hi Frank
I have ben trying to quell some upper mid harshness in my analogue (phono ) section with no success. I have messed with caps (types of0 and of course the rather voluminous jungle that is resistive loading.. so far i like the response tha 50 ohm loading has . The cartridge is ortofon mc20 Super mark11 (internal impedance of 5 ohms... ortofon recommend 20 to 150 ohms)
What i seem to be getting is a harshness on say violins, brass voices i reckon round 7k but could be having my self on re the frequency. It just isnt right.. if i play the cd of the same record the harshness is not there.. so heres the rub
What can i try?
i have tried the following
1.VTA
2.Cartridge alignment
3. Output caps on your head amp... i cant hear much difference on any value from .47uF to 3uF I have tried 4.7uf... The bass lost its definition at that value.
http://www.diyaudio.com/forums/atta...tamp=1179753229
Thats the link to the phono stage I hope Im not breaking forum rules by posting it here but i did download it from here.
I have read a lengthy froum post where you contributed .. and again i have become a tad lost in what can be done with a mc cart...
Any suggestions would be greatly appreciated.
Regards
Nick |
|
|
| fdegrove |
Hi,
I'm sure it's not the headamp.
From what you're describing it may be cartridge mistracking though.
As for the output cap, since this is prior to RIAA correction, the smaller value is usually the better but again, I don't think the problem lies with the electronics.
Moving about different TTs is often not possible but maybe you can move the electronics to someone else's place (assuming there's a well set up TT present) just to make sure?
Sorry if I can't be of more help, ;) |
|
|
| duderduderini |
Hi Frank
I know its not the headamp.. i am just out of ideas.. i reckon its the mc cart.... I might see if i can borrow one.
I wonder if its what Jim Hagermans' site refers to (the peak in the graphs).. You probably know of the page i speak of. The reason i have tried differing cap values is to see if that helped.
thanks for your help
Nick |
|
|
| duderduderini |
Hi Frank.
I got to thinking re your suggestion of cartridge mistracking..
I have been using a particular alignment guide all this time but i thought to get out the trusty ole ortofon guide i have... i was able to determine that the overhang I had been using was a tad on the short side (the ortofon guide has a 2 point stylus reference point and depending on wether the back of the cartridge is to the left or right of the marked parallel lines tells you wether to move the cartridge further out or in wrt to overhang)
So i was able to rid myself of the dreaded harshness.. tommorow i will tweak a bit more.. I mean whats is a day in my life when im not wearing my head mounted led light? (my lady calls it the geek light).
Thanks for the suggestion Frank.
You have legend status in my book
Regards
Nick |
|
|
|
