Hi all,
Kofi I´m currently using Valve Art 274Bs that I bought from Jac Music at a real fair price (something like 20 EURO each). Can´t tell the difference in sound with the chinese mesh 274As. 274A and 274B are the same tube with different base.
It is a 250-0-250 Transformer. I thought that it was OK to draw it like that...
Best regards,
Luis
Kofi I´m currently using Valve Art 274Bs that I bought from Jac Music at a real fair price (something like 20 EURO each). Can´t tell the difference in sound with the chinese mesh 274As. 274A and 274B are the same tube with different base.
It is a 250-0-250 Transformer. I thought that it was OK to draw it like that...
Best regards,
Luis
ligascon said:
I apologize if my comment regarding the transformer label seemed critical. That was not my intent. It's just that in your application, it is not a CT transformer. Just trying to help clear up some confusion. Experienced readers would have no problem with your label, and would recognize that it started out life at a CT.
Sheldon
Woo hoo! Glad that worked out-- saved me a bundle. I ordered two 274Bs today as well as the motor run caps for the PSU. I need to spread the cost over time, so that's all the buying I can do for a while.
I'm sure its OK-- I'm just a dumbass.
See?
Having a tough time getting Jack from Electra-print to respond to my request for a hand-wound PSU transformer, but I'm paitient. Any other winders you geeks would recommend?
Also, anything against using wirewounds for (what I think is) the voltage divider (with the paralleled 7K5 resistors) network in the audio circuit? Any what's the 100R pot for? Is this to adjust the fixed bias?
Wow. There's some whoppong dumb questions in here I'm sure. Go ahead. Give it to me. I can take it.
Kofi
Hi all,
It adjust the bias voltage for the d3A and hence the voltage supplied to the d3A and the 300B anodes. If you increase the bias voltage your d3A will work at a higher anode voltage and your 300b at a lower and the opposite. it sure comes in handy to balance things out...right now I´m running the d3A at about 2,11 Volts bias which gives me something like 175 Volts on the d3A anode and around 345 on the 300B. I like the sound this way.
Nothing that I can see. Just try to use a good quality metal film or tantalum for the 47K input and a carbon type for the d3A grid stopper (Bradley or Riken) The rest will make a difference but I don´t think it´s so critical.
Good luck and best whishes. Regards,
luis
It adjust the bias voltage for the d3A and hence the voltage supplied to the d3A and the 300B anodes. If you increase the bias voltage your d3A will work at a higher anode voltage and your 300b at a lower and the opposite. it sure comes in handy to balance things out...right now I´m running the d3A at about 2,11 Volts bias which gives me something like 175 Volts on the d3A anode and around 345 on the 300B. I like the sound this way.
Nothing that I can see. Just try to use a good quality metal film or tantalum for the 47K input and a carbon type for the d3A grid stopper (Bradley or Riken) The rest will make a difference but I don´t think it´s so critical.
Good luck and best whishes. Regards,
luis
Got it! Thanks!
Will do.
Well, you know... more questions. I'm really trying to make this a true learning experience, which will be trying for the DIYAudio community but I have faith in you(ze). So, in the interest of learning...
I thought that the 300b was supposed to see between a 3K and 4K plate load, but it looks to me like its seeing the 5.6K primary of the OPT and, therefore, a 5.6K load. Since matchematically, it must be looking at a 3.5K load to make 60mA at a 21V voltage drop, I know this is wrong, but I don't know why.
Also, I thought that the proper bias resistor to set the cathode voltage at 249V would have been about 5.8K, but the total resistance that the 300b cathode sees to ground looks to be about 4.8K. Once again, this is wrong, but I don't know why.
Also, my cousin has been touching me on the leg while nanny is away. I know this is wrong, but I don't know why.
Can ya help?
Kofi
Kofi Annan said:
Jack is old school. He's been very responsive on both my orders, but you have to call him. He doesn't respond as well to e-mail.
Sheldon
Kofi Annan said:
The transformer load has nothing to do with the bias current, which is DC, except for the DC resistance of the transformer. The load here refers to the reflected impedence - (primary turns/secondary turns)squared, times speaker impedence - for AC. I don't know much about ideal loads for 300b's but if your speakers are typical, they would have an actual impedence closer to 6 ohms, bringing your load for that transformer in the 3-4k range.
The 1k bias resistor would have a drop of 60v at 60 ma. This seems reasonable, as after that resistor, current is drawn off for the input tube's plate. This would indicate an input tube voltage of 249 - 60, or 189V. The actual voltage shown is 177 at the input plate, so the difference is probably in the resistance of the DRD choke. That current does not go through the paralleled 7.5k resistors, but reenters the stream below them. That current is shown as 15ma through the choke and 10ma through the input tube. They are in series and therefore must be the same, so one of those is wrong. No big deal, let's split the difference and call it 12.5V for convenience. So you have 60ma through the 1k for a drop of 60V, and 60-12ma, or about 48ma through the 3.75k (parallel 7.5k), for a drop of 180V. Now bring the two current streams together after the 3.75k and send the total 60ma through the 100 ohm resistor for 6V. So, 180 added to 60, plus 6, gets you 246. or close enough for tube work.
Can't help with the cousin though.
Sheldon
Very old school. I talked to him for about a half an hour this evening and he had lots of opinions about modifications to the DRD design, none very good, I'm afraid.
Anyway, he's going to wind me some power transformers and possibly a pair of OPTs, but I'll probably have to wait a while. He just got back from CES and he's pretty backlogged.
Wow! That's good information!
Even better! I think the 15mA on the DRD choke is a rating and not the actual amount of current drawn, but I probably have that wrong.
Thanks, man. Your description of the current flow really, really makes a lot of sense to me. Tell me something 5.2 million times and I'm likely to get it.
OK-- now for a really dumb one. Since this is a directly-heated triode, the filament voltage and the 1K resistor will both need to be connected to the same place, right? In other words, the filament voltage is tied to the proper pins (the two for the filament) and the cathode resistor is tied to one of the filament pins, right? Does it make a difference which of the filament pins the cathode resistor is tied to?
Hey-- the right answer gets you a date my cousin. She ain't a looker, but she knows where to touch!
Kofi
i'll touch the left leg of your cousin, i mean, the negative side...
no hum balance bot? no resistor pair? i know you're going dc but, it think it should still help.
ps. now how about your cousin? tell me more
pps. you're going DRD right? if you don't use hum pot or resistor pair, it will increase your bias voltage by 5V (filament voltage) and may screw up (a bit) all those computations above. but i won't get really anal about it
no hum balance bot? no resistor pair? i know you're going dc but, it think it should still help.
ps. now how about your cousin? tell me more
pps. you're going DRD right? if you don't use hum pot or resistor pair, it will increase your bias voltage by 5V (filament voltage) and may screw up (a bit) all those computations above. but i won't get really anal about it
Kofi Annan said:
You're right, that's very likely the case. I shoulda caught that.
Kofi Annan said:
That's what the schematic shows. It might make a little difference which side, but I think that' something you'd actually have to try. If I were doing it, I think I'd do as ArnoldC suggests and put a couple of 100ohm, 1watt resistors in series across the winding, and connect the 1k resistor between them. But maybe that's just being anal.
Sheldon
Hi people,
That´s correct. Sorry if it appears confusing on the schematic.
By the way Kofi, I´m trying to find Thorsten´s email where he details the ins an outs of the design...I´ll post it (if he doesn´t mind) I´m sure that it won´t anwser all of the questions but it will help you out understanding the whole thing.
Best regards,
Luis
That´s correct. Sorry if it appears confusing on the schematic.
By the way Kofi, I´m trying to find Thorsten´s email where he details the ins an outs of the design...I´ll post it (if he doesn´t mind) I´m sure that it won´t anwser all of the questions but it will help you out understanding the whole thing.
Best regards,
Luis
Kofi Annan said:
Yes. With the resistor on one end only there will be a slight difference in potential from one end of the filament to the other. With the resistor divider, the filament will vary from middle to each end, or balanced, and the voltage delta from highest to lowest on the filament will be 1/2. Does it matter? Don't know.
Kofi Annan said:
It doesn't have to be too exact. Err on the high side. You can always drop a little voltage with series resistance.
Sheldon
Makes sense to me.
Will do, but I now have a bit of a dilemma. Jack Eliano took a look at the schematic for the modified DRD and is not thrilled about it. He seems to think that there are some issues with the design and is advising that I go with the original DRD circuit he has posted on his site.
Jack seems very honest and I don't think he'd try and steer me wrong, but I also know that he's quite opinionated and is bit upset that someone has toyed with the original DRD design.
Now, I'm just trying to play the middle here and I don't want to make any waves, but his opinion (his opinion) is that the Thorsten design is far too complicated in the PSU and has some unnecessary additions in the audio circuit that could cause some instability.
Now, I am of the "if it sounds good..." camp, so I'd be willing to bet that both of these sound great, but I am also trying to learn something here, so let me make available the "Jack" side of the equation.
--THE OPINIONS ESPOUSED BELOW DO NOT NECESSARILY REPRESENT THOSE OF STAFF AND MANAGEMENT AT THE UNITED NATIONS, PARTICULARLY NOT NECESSARILY THOSE OF ONE KOFI ANNAN--
He first mentioned that he thought this circuit would cause a few problems as the bias point for the 300b and the D3a would be feed another. I believe what he means is that if you just want to adjust the bias point of the 300b, you're going to end up adjusting the bias point of the D3a as well. Moreover, he thought that the driver tube should have about 15mA and 200V to eliminate the Miller effect on the 300b at frequencies below 20KHz.
You can tell these ain't Kofi's words, right?
Here's just some of the rants I was able to scribble down. Jack sez:
1. The common mode chokes in the PSU are only for filtering RF frequencies and are unnecessary, unless you live near an arc welding factory.
2. Elevating the D3a filament by strapping it to the B+ is also unnecessary.
3. The paralleling of caps coming off the 300b cathode are unnecessary-- no audible difference between one 50uF - 100uF cap (depending on desired bass roll-off) and the multiple paralleled nanofarad caps.
4. The parallel caps coming off the choke load can be done away with entirely with the other end of the choke strapped to the top of the paralleled 7K5 resistors.
5. Given scenario 4, the 300b should be biased with a 5K variable resistor and adjusted so that it sees 220V and 85mA across it. The resistor network coming off the 300B cathode can then be dispensed with.
In a nutshell, I think Jack was concerned that there were modifications made to a circuit that had his name on it, although he was quick to mention that the DRD circuit is not really new--- it just a modified Lofton-White (I'm sounding studious here, I know). The important thing for me is that I really, really want to learn. I learn most by doing and I would like to get this amp project rolling.
I am compelled by what Jack had to offer and I think I could learn a lot by asking for comments on his comments. I know this will likely ruffle some feathers and I don't mean to do that. Hell, I don't really understand half of what he said, but I know this will make for some interesting discussion, so here it is in print.
So, have at it. Let me know what you think of Jack's comments. But remember, that I'm just a schmuck who's trying to sort all this out.
Jeez. This makes the cousin thing seem trivial.
Thine,
Kofi
This ain't so hard. Build it Jack's way first, since that's simpler. Then, as you have the time and inclination modify one element at a time (adding a common mode choke, for instance) and see if you like it better or not, etc.. I didn't do a careful analysis, but I don't think anything in the original design forecloses the option of making all the mods to the posted design at very little additional cost. This way you will learn the most important thing - what YOU like.
Or, just ignore Jack and build whatever you want. BTW, he tried to talk me out of my present project, but I'm keen to try it anyway. Still he's fun to talk to.
Sheldon
Or, just ignore Jack and build whatever you want. BTW, he tried to talk me out of my present project, but I'm keen to try it anyway. Still he's fun to talk to.
Sheldon
Hi people,
Yes Kofi, things get complicated and then you start whishing that your cousin was around...
I wish I could anwser to the issues that Jack brings to the table but as you could have probably guessed, I don´t have the background.
I only built an amp following the ideas of someone that showed himself totally helpful and then I shared the process with you guys because the thing has sounded great since day 1 and continues to do so without a glitch.
Maybe big brother T. himself is watching how this whole thing unveils. Also, maybe is time that he could bring some discurssion items to our table...
As for now I will take a chance and quote some of the info that he gave me about the amp. Hope that it helps. Here it is:
Hope that you find it helpful. Take much care,
Luis
Yes Kofi, things get complicated and then you start whishing that your cousin was around...
I wish I could anwser to the issues that Jack brings to the table but as you could have probably guessed, I don´t have the background.
I only built an amp following the ideas of someone that showed himself totally helpful and then I shared the process with you guys because the thing has sounded great since day 1 and continues to do so without a glitch.
Maybe big brother T. himself is watching how this whole thing unveils. Also, maybe is time that he could bring some discurssion items to our table...
As for now I will take a chance and quote some of the info that he gave me about the amp. Hope that it helps. Here it is:
Hope that you find it helpful. Take much care,
Luis
Hi,
A bit of history (yes, I know that is HisStory).
My Circuit actually grew seperate to the DRD and parallel (check the Joe List archives). Hence my circuit is NOT a modified DRD. It is a seperate complete development.
I have of course since seen the DRD. I think that Jack Elliano is looking for different qualities and has different design approaches. An example is that he invarably runs output valves at max dissipation, something I fiond to generate much worse sound than possible from the output valve.
I suggest that you stick one design. By all means build Jacks circuit, but his COMPLETE CIRCUIT EXACTLY as he shows. Or build mine and then EXACTLY as I suggest.
As you have seen it fit to make Jacks criticsm of my circuit public I will answer the points raised.
Jack Eliano took a look at the schematic for the modified DRD and is not thrilled about it. He seems to think that there are some issues with the design and is advising that I go with the original DRD circuit he has posted on his site.
He is wrong. The circuit contains the same positive feedback operating condition stabilisation as found in old Loftin White circuit. Unlike the DRD circuit the "Monkey" is not subject to the pinching off of anode current due to rise of cathode voltage when driven into high distortion. Hence the "instability" is actually an IMPROVED stability, over most other direct coupled circuits.
The 300B operating point in my circuit is self adjusting, it SHOULD NOT be adjusted manuual.
Funny, his OWN circuit operates the input/driver valve at much LESS anode current than the D3a is operated in my circuit, so low current is good when he use it?
The difference is actually visible on a 'scope and noise is not just generated by arc welding factories, but also by rectifiers, switched mode supplies, mobile phones, whashing machines, fridges and other modern conveniences...
He is wrong. I have found more then a few cases where individual samples of valves showed significant heater/cathode noise leakage, meaning very unusal hum pattern where observed when these valves where AC heated.
By elevating the heater of ANY AC heated valve (I even fo this for EL34's and KT88's in my modified Shanling Amplifiers) the parasitic heater/cathode diodes are reilably reverse biased, which reliably produces low noise with AC Heating AND reduced non-linear signal current leakage. I found the result of this little "tweak" quite significant in sonic terms.
Actually, the paralleled capacitors perform what is called bypassing. Look it up. This is a contentious issue, having tried both options I ALLWAYS use bypass capacitors. Several people who buyild my designs and did the same comparison for themselves agree. Non who have ACTUALLY TRIED IT have said it made no difference.
Then the circuit operation is altered in a way that reduces LF and HF bandwidth. I must councel AGAINST removing these capacitors in the strongest possible terms. It is in fact the WE Style decoupling found in the DRD Output stage and re-named by Jack "Ultrapath".
That will result in an Elliano DRD Amplifier.
I personally do not think that either the operating point for the 300B offers that good sound quality. It is worth noting the Ron Welborne's DRD Kits also use a lower dissipation opertaing point for better sound. Also, because it is an operating condition at the limit of dissipation the adjustment is needed to avoid exceeding the valves maximum dissipation for valves at the edge of tolerance are not killed quickly.
My circuit does not require adjustment other than for the D3a tolerance as the Output Valve is operated under condition that give less power but in stead are optimised for best sound from the 300B.
Actually, that Circuit is NEITHER a modification of Jack's circuit nor does it (or shouyld it have) Jacks name on it.
I do not think that his criticisms where based on actually trying to un derstand why I had made the circuit design as I have, but rather to assert that the Circuit should be like his circuit.
Allow me to assert that my circuit should be like my circuit, as I have very good reasons for each and every feature in it.
Sayonara
Kofi Annan said:
A bit of history (yes, I know that is HisStory).
My Circuit actually grew seperate to the DRD and parallel (check the Joe List archives). Hence my circuit is NOT a modified DRD. It is a seperate complete development.
I have of course since seen the DRD. I think that Jack Elliano is looking for different qualities and has different design approaches. An example is that he invarably runs output valves at max dissipation, something I fiond to generate much worse sound than possible from the output valve.
I suggest that you stick one design. By all means build Jacks circuit, but his COMPLETE CIRCUIT EXACTLY as he shows. Or build mine and then EXACTLY as I suggest.
As you have seen it fit to make Jacks criticsm of my circuit public I will answer the points raised.
Jack Eliano took a look at the schematic for the modified DRD and is not thrilled about it. He seems to think that there are some issues with the design and is advising that I go with the original DRD circuit he has posted on his site.
Kofi Annan said:
He is wrong. The circuit contains the same positive feedback operating condition stabilisation as found in old Loftin White circuit. Unlike the DRD circuit the "Monkey" is not subject to the pinching off of anode current due to rise of cathode voltage when driven into high distortion. Hence the "instability" is actually an IMPROVED stability, over most other direct coupled circuits.
Kofi Annan said:
The 300B operating point in my circuit is self adjusting, it SHOULD NOT be adjusted manuual.
Kofi Annan said:
Funny, his OWN circuit operates the input/driver valve at much LESS anode current than the D3a is operated in my circuit, so low current is good when he use it?
Kofi Annan said:
The difference is actually visible on a 'scope and noise is not just generated by arc welding factories, but also by rectifiers, switched mode supplies, mobile phones, whashing machines, fridges and other modern conveniences...
Kofi Annan said:
He is wrong. I have found more then a few cases where individual samples of valves showed significant heater/cathode noise leakage, meaning very unusal hum pattern where observed when these valves where AC heated.
By elevating the heater of ANY AC heated valve (I even fo this for EL34's and KT88's in my modified Shanling Amplifiers) the parasitic heater/cathode diodes are reilably reverse biased, which reliably produces low noise with AC Heating AND reduced non-linear signal current leakage. I found the result of this little "tweak" quite significant in sonic terms.
Kofi Annan said:
Actually, the paralleled capacitors perform what is called bypassing. Look it up. This is a contentious issue, having tried both options I ALLWAYS use bypass capacitors. Several people who buyild my designs and did the same comparison for themselves agree. Non who have ACTUALLY TRIED IT have said it made no difference.
Kofi Annan said:
Then the circuit operation is altered in a way that reduces LF and HF bandwidth. I must councel AGAINST removing these capacitors in the strongest possible terms. It is in fact the WE Style decoupling found in the DRD Output stage and re-named by Jack "Ultrapath".
Kofi Annan said:
That will result in an Elliano DRD Amplifier.
I personally do not think that either the operating point for the 300B offers that good sound quality. It is worth noting the Ron Welborne's DRD Kits also use a lower dissipation opertaing point for better sound. Also, because it is an operating condition at the limit of dissipation the adjustment is needed to avoid exceeding the valves maximum dissipation for valves at the edge of tolerance are not killed quickly.
My circuit does not require adjustment other than for the D3a tolerance as the Output Valve is operated under condition that give less power but in stead are optimised for best sound from the 300B.
Kofi Annan said:
Actually, that Circuit is NEITHER a modification of Jack's circuit nor does it (or shouyld it have) Jacks name on it.
I do not think that his criticisms where based on actually trying to un derstand why I had made the circuit design as I have, but rather to assert that the Circuit should be like his circuit.
Allow me to assert that my circuit should be like my circuit, as I have very good reasons for each and every feature in it.
Sayonara
KYW,
most probably you expressed your thougths, ideas and experience several times in the past but unfortunately I missed your older posts on the subject. I had also a look to a link to a thread from yourself posted by Luis.
I would ask what are the pros in using a pentode as driver in this project. ... references to exisiting literature would be great.
Ciao. Grazie.
Gianluca
most probably you expressed your thougths, ideas and experience several times in the past but unfortunately I missed your older posts on the subject. I had also a look to a link to a thread from yourself posted by Luis.
I would ask what are the pros in using a pentode as driver in this project. ... references to exisiting literature would be great.
Ciao. Grazie.
Gianluca
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.