Re: Re: Re: Re: Re: Re: Re: Re: mods.
plug this part number into the digi-key search engine:
DSEI8-06A-ND
it's a FRED...fast recovery epitaxial diode.
if you want to totally max out that little amp, your best bet would be tube recitification, choke, smaller caps, and triode/no NFB bridging. also replace plate load resistors with carbon film or tants...www.angela.com
compare THAT to a tripath on a sensitive full ranger!
planet10 said:
plug this part number into the digi-key search engine:
DSEI8-06A-ND
it's a FRED...fast recovery epitaxial diode.
if you want to totally max out that little amp, your best bet would be tube recitification, choke, smaller caps, and triode/no NFB bridging. also replace plate load resistors with carbon film or tants...www.angela.com
compare THAT to a tripath on a sensitive full ranger!
Upgrades and Modifiactions
Greetings,
So I tried out some of the VoltSecond modifications. Results were good. You can read about them on my webpage.
I have read that a choke did wonders on this little amp. I will keep that in mind, but fitting a choke onto my enclosure will be tough. Tube rectification would be nice, but I don't have the real estate.
BTW - I replaced the bridge with 4 UF4007.
As for the resistors, they are already carbon film with the exception of the power supply resistors which are metal oxide. Another member of the form tried resistor upgrades but did not find much of an improvement.
http://www.diyaudio.com/forums/showthread.php?postid=227912#post227912
Tants will cost almost as much as the kit and premium carbon film such as Riken are also pricey at $3-4 a pop. Used carbon comps could be implimented at salvage costs, but is replacing generic 5% carbon films with vintage 10-20% carbon comps a real improvement? Heat may be a problem with vintage carbon comps.
Further upgrades that I am contemplating:
Removing the DC blocking caps - as suggested by Blackie
Adding MKP caps before and after the bridge.
Upgrading the poly coupling caps with nice film-foil caps.
Cheers,
Gio.
noisenyc said:
Greetings,
So I tried out some of the VoltSecond modifications. Results were good. You can read about them on my webpage.
I have read that a choke did wonders on this little amp. I will keep that in mind, but fitting a choke onto my enclosure will be tough. Tube rectification would be nice, but I don't have the real estate.
BTW - I replaced the bridge with 4 UF4007.
As for the resistors, they are already carbon film with the exception of the power supply resistors which are metal oxide. Another member of the form tried resistor upgrades but did not find much of an improvement.
http://www.diyaudio.com/forums/showthread.php?postid=227912#post227912
Tants will cost almost as much as the kit
and premium carbon film such as Riken are also pricey at $3-4 a pop. Used carbon comps could be implimented at salvage costs, but is replacing generic 5% carbon films with vintage 10-20% carbon comps a real improvement? Heat may be a problem with vintage carbon comps.Further upgrades that I am contemplating:
Removing the DC blocking caps - as suggested by Blackie
Adding MKP caps before and after the bridge.
Upgrading the poly coupling caps with nice film-foil caps.
Cheers,
Gio.
Re: Upgrades and Modifiactions
those carbon films should sound fine. wirewounds might be nice in the power supply, mouser has some for not too much dough.
GG said:
Greetings,
So I tried out some of the VoltSecond modifications. Results were good. You can read about them on my webpage.
I have read that a choke did wonders on this little amp. I will keep that in mind, but fitting a choke onto my enclosure will be tough. Tube rectification would be nice, but I don't have the real estate.
BTW - I replaced the bridge with 4 UF4007.
As for the resistors, they are already carbon film with the exception of the power supply resistors which are metal oxide. Another member of the form tried resistor upgrades but did not find much of an improvement.
http://www.diyaudio.com/forums/showthread.php?postid=227912#post227912
Tants will cost almost as much as the kit
Further upgrades that I am contemplating:
Removing the DC blocking caps - as suggested by Blackie
Adding MKP caps before and after the bridge.
Upgrading the poly coupling caps with nice film-foil caps.
Cheers,
Gio.
those carbon films should sound fine. wirewounds might be nice in the power supply, mouser has some for not too much dough.
Heh, I'm new to tube amps and have been lurking diyAudio.com until today. Hey, hello.
I received and assembled my K-12M on Wednesday night. For lack of parts, this is a stock kit amp, excepting that I have foregone the plastic power switch. I'm pretty pleased with the overall sound; it compares favorably with my Alesis RA-100 solid state amp on my SPL 94dB 1w/1m loudspeakers.
Within a few feet of the midrange/tweeter drivers, I notice some 60hz harmonics buzzing quietly. I intend to try the AC snubber and UF diode upgrades this weekend if the local radio shack can give me sweet instant gratification...
I am a bit concerned that one of the four 11MS8 tubes' plates can be seen dimly glowing in one corner if I turn off the lights and let my eyes adjust. After several combinations of tube ordering, I can say with a fair degree of certainty that the problem is following the tube...
I'm wondering to what extent the 11MS8s can be matched. A rep at S-5 Electronics states "our testings show that they are all very close....at least for the kit." Hrm.
Rk for the power pentode push-pull pairs (R8, R17 - 300ohm, 3w) drops about 24.5 and 25.4v DC respectively, or 80mA and 83.3mA cathode current for the two pairs (one of which contains a glowing tube, which seems to follow the tube). The pentode plate voltage was 204 VDC on all four tubes, so the voltage from cathode to plate on the tubes is up around 180-181VDC.
The tube data ( 11MS8 ) for the RCA version of this tube says the zero-signal plate current should be 50 Ma at 120V as a class A amplifier. Now... this is a class AB1 push-pull amplifier with quite a bit more plate voltage...
I received and assembled my K-12M on Wednesday night. For lack of parts, this is a stock kit amp, excepting that I have foregone the plastic power switch. I'm pretty pleased with the overall sound; it compares favorably with my Alesis RA-100 solid state amp on my SPL 94dB 1w/1m loudspeakers.
Within a few feet of the midrange/tweeter drivers, I notice some 60hz harmonics buzzing quietly. I intend to try the AC snubber and UF diode upgrades this weekend if the local radio shack can give me sweet instant gratification...
I am a bit concerned that one of the four 11MS8 tubes' plates can be seen dimly glowing in one corner if I turn off the lights and let my eyes adjust. After several combinations of tube ordering, I can say with a fair degree of certainty that the problem is following the tube...
I'm wondering to what extent the 11MS8s can be matched. A rep at S-5 Electronics states "our testings show that they are all very close....at least for the kit." Hrm.
Rk for the power pentode push-pull pairs (R8, R17 - 300ohm, 3w) drops about 24.5 and 25.4v DC respectively, or 80mA and 83.3mA cathode current for the two pairs (one of which contains a glowing tube, which seems to follow the tube). The pentode plate voltage was 204 VDC on all four tubes, so the voltage from cathode to plate on the tubes is up around 180-181VDC.
The tube data ( 11MS8 ) for the RCA version of this tube says the zero-signal plate current should be 50 Ma at 120V as a class A amplifier. Now... this is a class AB1 push-pull amplifier with quite a bit more plate voltage...
chromal said:
howdy again!
sorry, i was in a bit of a hurry when i made my last post.
check the manual for the design max plate dissipation.
then just do the math to make sure the design maximum current is not exceeded. you are totally on the right track. calculate current per tube and multiply X plate volts for dissipation in watts.
generally speaking, well made NOS tubes are okay if the plate voltage max is a bit exceeded, sometimes a LOT exceeded, as long as the overall dissiaption is within design maximums. glowing plates is generally not groovy on AB pentodes tho...they will gas up and/or fail prematurely....
you can install a bias balance pot if you wish so that unmatched tubes wont be an issue. you will find with cathode bias arrangements there is a fair amount of leeway on the bias resistor. bigger resistor=less current draw but plate voltages will tend to rise. even more so on tube rectified circuits. sound will change...many tubes sound good low'n'hot...depends on the tube and the loading...
have fun!
An externally hosted image should be here but it was not working when we last tested it.
There's a low-light shot of the tube overheating. It's not really visible in normal light, but with a 10 second exposure on the digital camera, you can see it clearly. (The right pair is also dimly glowing, though this isn't really evident to the naked eye).
Okay; the RCA data source says the pentode max ratings are 250V 6W. No input/Idle measurments: 25VDC at cathode at 83.3mA current across 300ohm resistor for both tubes together (screen current neglected), 205VDC at plates makes for a 180v differential between cathode and plate. So... 180V * 83.3mA = 14.994w, or about 7.5w per plate at 205vdc...
Does this mean that the pentodes are dissipating 1.5w beyond the max rated 6w spec *on average*? (with the understanding that a common cathode current between two pentode cathods in a push-pull arrangement will not say a thing about indivudual pentode currents.)
I've been reading a bit about biasing here. I guess that the practical offshot of all this is that it's sounding as if I might want to consider a modification that allows me to vary cathode resistor's impendence independantly per tube, perhaps the cathode 'balance' potted approach you mentioned...
chromal said:
that tube is def running too hot.
when you do the calculations, use the differential plate/cathode figure, that is the true plate voltage.
otherwise, your calculations are correct.
you can also measure the actual current running thru individual tubes if you have a meter, and check matching.
set meter on Ma setting, and connect one lead to the plate, the other lead to the center tap of the output trannie.
there are a few choices of bias schemes, each will produce different sonics.
matching between tubes should be within 5 ma any way you shake it.
if you get yourself a Radiotron Designer's Handbook, 3rd edition, you have all the bases covered. pretty much everything is covered in there, except for some fun esoterica. all the basics. a fun read for the enthusiast.
upgraded caps
FWIW, I also built this K502 kit a few months ago and upgraded the capacitors to four Jensen PIO caps and 2 Jupiter beeswax capacitors and think it imparts a wonderful sound. I also am very impressed with this amp, which is my first tube amp. It is amazing to go back and listen to old music and hear the dfference in quality through a tube amp. I find myself listening to stuff I hven't heard in a long time just to hear new nuances in the music. Your chassis looks like it turned out pretty well.
FWIW, I also built this K502 kit a few months ago and upgraded the capacitors to four Jensen PIO caps and 2 Jupiter beeswax capacitors and think it imparts a wonderful sound. I also am very impressed with this amp, which is my first tube amp. It is amazing to go back and listen to old music and hear the dfference in quality through a tube amp. I find myself listening to stuff I hven't heard in a long time just to hear new nuances in the music. Your chassis looks like it turned out pretty well.
Okay. I decided to try to measure plate current. I wasn't sure how much screen current was contributing to the cathode current and I especially wanted to see what the individual tubes' currents were.noisenyc said:
b+ to P1 across the output transformer read 83.5ohm, b+ to P2 on the output transformer read 74.2ohm. Both output transformers basically conformed to this.
b+ voltage at the transformer read 210VDC. On no-signal, P1 read as follows (pin 6 = pentode plate):
U1,6: 207.6V (-2.4V) => 2.4V/74.2o = 32.3mA \ -- 8.4mA diff
U2,6: 206.6V (-3.4V) => 3.4V/83.5o = 40.7mA /
U3,6: 207.2V (-2.8V) => 2.8V/74.2o = 37.7mA \
U4,6: 205.1V (-4.9V) => 4.9V/83.5o = 56.3mA / -- 18.6mA diff (!)
U1/U2 pentode plate current sum: 73.0mA \ -- 9.4mA diff (screen current?)
U1/U2 pentode cathode current: 82.4mA /
U3/U4 pentode plate current sum: 96.3mA \
U3/U4 pentode cathode current: 86.1mA / -- 10.2mA diff (screen current?)
So, given these numbers, it shouldn't be surprising that U4's pentode was, indeed, the one overheating. So, let's see...
Cathode voltage was shared between U3/U4 @ 25.83V. U4's pentode at 205.1V, so 205.1-25.83 = 179.3V. 179.3V*56.3mA = 10w!
Running the numbers:
U4 = 10.1w
U3 = 6.8w
U2 = 7.4w
U1 = 5.9w
Tube's max dissipation is 6w. U2, U3, and especially U4 are exceeding max dissipation specs, and on zero signal, too.
Or I made a mistake somewhere.
That's not unlikely, but the numbers 'track' the hottest tube... Those 3w 300ohm cathode resisters are sure getting a workout at zero signal, too...
chromal said:
good move. a couple of suggestions/observations...
this is always the case with p/p trannies. the two halves will measure a different series resistance.
i suggest you measure individual tube current draw using the "shunt" method as described in the aikenamps link lifted from the text of aspen pittman's book. it's quite accurate.
also, screen current draw is basically a non-issue in the quiescent state. most cheapo amps ignore the screen ratings of pentodes and simply use a series resistor to limit screen current when the tube is driven. that pretty much sucks but it will work if the series resistor is big enough. most common power pentodes have a screen *voltage* rating considerably below the plate voltage rating. if you are designing from the ground up, it's good to follow the guidelines as it improves sound considerably....but then, triode strapping/nfb removal (see my earlier posts) makes a quantum improvement in *most* cheapo pentode amps...
as you can see, it would be a simple matter to convert this amp to a class A P/P "triode" amp.
So, given these numbers, it shouldn't be surprising that U4's pentode was, indeed, the one overheating. So, let's see...
Cathode voltage was shared between U3/U4 @ 25.83V. U4's pentode at 205.1V, so 205.1-25.83 = 179.3V. 179.3V*56.3mA = 10w!
Running the numbers:
U4 = 10.1w
U3 = 6.8w
U2 = 7.4w
U1 = 5.9w
Tube's max dissipation is 6w. U2, U3, and especially U4 are exceeding max dissipation specs, and on zero signal, too.
Or I made a mistake somewhere.
3 watts is pretty marginal for a power tube cathose resistor.
you could halve the cathode resistance and bias each tube separately. this will self balance *somewhat*.
you could use a series pot, connecting the legs to the cathodes and the wiper to the cathode resistor and tweak the balance. just make sure the whole series resistance yields an overall current draw below the design maximum. and use higher wattage parts fer chrissake haha the dale and vishay wirewounds at mouser are good.
you could start with a larger stock of tubes and pop 'em in until you find ones that are happy in stock circuit.
just remember that the design maximums are quiescent ratings, and the driven ratings are basically allowed for.
and you can cheat voltages somewhat with good tubes.
My K-12M's doing the same thing
I followed a link to this forum from the Audio Asylum forums. I've been trying different tubes like crazy to try to eliminate the glowing plate problem on my K-12M. I'm not experienced in circuit design, so I figured I'd just ask if someone else has already done some kind of modification to the K-12M circuit to give it the gift of individual per-tube pentode biasing. I've almost spent more than the cost of the kit itself on 11MS8 tubes, trying to get two pairs that didn't have plates you could roast marshmellows next to.
There's only one tube store local to me and I already own his best two pentode matched pairs. I also bought four on eBay which were matched horribly. 
That's the best I've managed - a pair of Panasonics which don't glow (and sound amazing!) and a pair of Dumonts which sound so-so (probably due to being mismatched) and one glows. That's my reward for building a nice home for this amp out of a gutted Hughes DirecTV box. (link)
I followed a link to this forum from the Audio Asylum forums. I've been trying different tubes like crazy to try to eliminate the glowing plate problem on my K-12M. I'm not experienced in circuit design, so I figured I'd just ask if someone else has already done some kind of modification to the K-12M circuit to give it the gift of individual per-tube pentode biasing. I've almost spent more than the cost of the kit itself on 11MS8 tubes, trying to get two pairs that didn't have plates you could roast marshmellows next to.
There's only one tube store local to me and I already own his best two pentode matched pairs. I also bought four on eBay which were matched horribly.
An externally hosted image should be here but it was not working when we last tested it.
That's the best I've managed - a pair of Panasonics which don't glow (and sound amazing!) and a pair of Dumonts which sound so-so (probably due to being mismatched) and one glows. That's my reward for building a nice home for this amp out of a gutted Hughes DirecTV box. (link)
noisenyc said:
I appreciate the pointer! Looks to be out of print, though if I see a cheap copy I'll be sure to grab it. Yeesh, even fetching decent prices on ebay. Maybe that's one I'll find in the library. My copy of Morgan Jones' Building Tube Amplifiers arrived today, though its focus seems to be on construction praxis rather than circuit theory. Which is fine, too...
Aikenamps seems to be a nice treasure trove of nice papers. Good lead, thanks!
I ran the shunt method, seeing as my Fluke 73III could do it and all... I guess I didn't see the value in shunting around the transformer, since its windings must still be passing current... I thought that would throw my readings unless I opened the circuit and placed the meter inline... So, heh, I'm not sure how good these results are, but DMM shunt says:
U1: 37.4mA @ 178.8V - 6.9w
U2: 36.8mA @ 178.6V - 6.6w
U3: 46.9mA @ 176.5V - 8.3w
U4: 32.1mA @ 177.3V - 5.7w
(All plate voltages given above are between the plate and the cathode of the tube in question). Oh, and, yes, I reordered the tubes last night in an attempt to reduce the zero signal current. U3 is now the 'hot' tube.
Yeah, so simple that I've been tempted to try. The amp definitely has more output power than my SPL 94dB 1w/1m loudspeakers need for my purposes, though it's a bit soft in the low-end freqs. Some folks have been upgrading the power caps as a means of realizing less low freq attenuation, which sound reasonable to this newbie.
yeah... any reason not to go something more like 10w?
I'm definitely still attracted to the idea of a cathode balance pot for cathode biasing... I want to understand the potential pitfalls a little better before I proceed. But... if I could literally just dial in the currents to match... I wonder if two pots in series-- one for the balance and one to move the shared resistance around. Of course, I guess that at that point, individual tube pots would be more sensible? Or not? I don't think I've seen a schematic with this arrangement, I'm sure there's a good reason...
YEah, I'm working on this, actually. But making the amp handle tube mismatches in the long term definitely has a certain appeal.
Hrm, well, that definitely explains a little of my confusion. So, the maxes are given with the expectation of being exceeded part of the AC cycle. . . Some data sheets are more clear about this than others. I'd love to see a better datasheet, maybe even with characteristic curves, for the 11MS8. I wonder who made the tubes S-5 has been including with thier kit...
[/QUOTE]
Funny you should mention the shunt method, I just tried the exact same thing (Hey, why not live dangerously?) and here's what I got:
(Starting at the tube next to the electrolytic caps as "Tube 1")
Stock "Made In Japan" 11MS8 tubes:
Cathode v = 183.5
Tube 1: 40ma * 183.5 = 7.34w
Tube 2: 39.2ma * 183.5 = 7.1932w
Cathode v = 182.6
Tube 3: 40.1ma * 182.6 = 7.32226w
Tube 4: 40.5ma * 182.6 = 7.3953w
Panasonic / Dumont 11MS8 tubes:
Cathode v = 187.2
Panasonic
Tube 1: 35.4ma * 187.2 = 6.62688w
Tube 2: 36.6ma * 187.2 = 6.85152w
Cathode v = 185.3
Dumont
Tube 3: 35ma * 185.3 = 6.4855w
Tube 4: 44.3ma * 185.3 = 8.20879w (Hello, my hot little friend!)
Now, from what I gathered from that page about biasing a class AB push-pull amp, it said it needs to be biased at 70% idle dissipation (6w / 0.7 = 4.2 watts). I'm guessing the proper bias for this amp on my mains voltage would be in the neighborhood of 22ma (or 32ma assuming there's some strange need for it to be biased at 100% idle dissipation).
I guess the designer of this amp figured the maximum plate dissipation ratings were more like guidelines than rules. The only thing I can think of is that this amp is the same as the 11BM8 kit that AES sells, but with a different tube pinout. The 11BM8 has a 7 watt plate dissipation, which puts it right in the ballpark.
Now I'm trying to learn as I go here, so if I made some mistakes feel free to correct me. I have a basic understanding of electronics but this tube stuff is new to me. I've been looking at some schematics of other tube amps and noticed that the bias current on an amp like my Jolida is supplied by a seperate winding on the power transformer. It looks like the only way to do individual biasing would be to get rid of the two (300 ohm, 3 watt) stock cathode resistors, cut the trace connecting the cathode pairs, then run a new, slightly lower value (but still high wattage) resistor and pot in series between ground and each tube's pentode section cathode. If this sounds about right, any recommendations for the resistor and pot?
I'm not really interested in removing the negative feedback and triode strapping since I wish this amp had a bit more oomph as it is. It's somewhat depressing to find out that as it stands, it's running its tubes far too hot.
(Starting at the tube next to the electrolytic caps as "Tube 1")
Stock "Made In Japan" 11MS8 tubes:
Cathode v = 183.5
Tube 1: 40ma * 183.5 = 7.34w
Tube 2: 39.2ma * 183.5 = 7.1932w
Cathode v = 182.6
Tube 3: 40.1ma * 182.6 = 7.32226w
Tube 4: 40.5ma * 182.6 = 7.3953w
Panasonic / Dumont 11MS8 tubes:
Cathode v = 187.2
Panasonic
Tube 1: 35.4ma * 187.2 = 6.62688w
Tube 2: 36.6ma * 187.2 = 6.85152w
Cathode v = 185.3
Dumont
Tube 3: 35ma * 185.3 = 6.4855w
Tube 4: 44.3ma * 185.3 = 8.20879w (Hello, my hot little friend!
)Now, from what I gathered from that page about biasing a class AB push-pull amp, it said it needs to be biased at 70% idle dissipation (6w / 0.7 = 4.2 watts). I'm guessing the proper bias for this amp on my mains voltage would be in the neighborhood of 22ma (or 32ma assuming there's some strange need for it to be biased at 100% idle dissipation).
I guess the designer of this amp figured the maximum plate dissipation ratings were more like guidelines than rules.
The only thing I can think of is that this amp is the same as the 11BM8 kit that AES sells, but with a different tube pinout. The 11BM8 has a 7 watt plate dissipation, which puts it right in the ballpark.Now I'm trying to learn as I go here, so if I made some mistakes feel free to correct me. I have a basic understanding of electronics but this tube stuff is new to me. I've been looking at some schematics of other tube amps and noticed that the bias current on an amp like my Jolida is supplied by a seperate winding on the power transformer. It looks like the only way to do individual biasing would be to get rid of the two (300 ohm, 3 watt) stock cathode resistors, cut the trace connecting the cathode pairs, then run a new, slightly lower value (but still high wattage) resistor and pot in series between ground and each tube's pentode section cathode. If this sounds about right, any recommendations for the resistor and pot?
I'm not really interested in removing the negative feedback and triode strapping since I wish this amp had a bit more oomph as it is. It's somewhat depressing to find out that as it stands, it's running its tubes far too hot.
ha now you guys are cookin.
here's a few points.
the radiotron is available on the web somewhere in reprint on CD-ROM. theres a lot of decent sources but this is the BIBLE. it covers most aspects of tube circuit design....try "audio amateur" in your google.
individual pots on cathodes for bias are rarely seen due to the prohibitive cost of 4 higher wattage pots per amp. do the math to determine the necessary wattage to handle the current you are running thru the load (resistor). ohm's law again! darn him! in my designs i multiply the minimum wattage necessary by at least 4.
it's heat, lads, and heat is murder. hot resistors drift, fail, etc.
grounding the cathodes and building a negative supply to provide bias is certainly an option. you could add a trannie, etc to get the voltage, or derive the voltage in various ways from the existing supply. if you study schematics avaliable at various sites on the web you will see some clever schemes. if you choose this route you will get back the voltage lost thru the cathode resistors and voila, more power. but...will that SOUND better?
consider this. you can also lower the plate voltage to bring things into line. low and hot, low and hot. yeah baby. you will note that the class "A" ratings in the tube manuals generaly show plate voltages well BELOW the maximum...because these tube designers were telling you something. run the tubes in the sweet spot on the curves for linearity, good sound.
look at the hammond chokes at angela.com. they got some DC resistance, hint hint.
now you have a perfectly good reason to put a choke in this supply. or even a tube rectifier. don't be afraid to cut traces and run wires to mod this thang.
you boys will learn much by studying the traces left by the ancient ones hehe. the radiotron, old schematics, and tube manuals.
here's a few points.
the radiotron is available on the web somewhere in reprint on CD-ROM. theres a lot of decent sources but this is the BIBLE. it covers most aspects of tube circuit design....try "audio amateur" in your google.
individual pots on cathodes for bias are rarely seen due to the prohibitive cost of 4 higher wattage pots per amp. do the math to determine the necessary wattage to handle the current you are running thru the load (resistor). ohm's law again! darn him! in my designs i multiply the minimum wattage necessary by at least 4.
it's heat, lads, and heat is murder. hot resistors drift, fail, etc.
grounding the cathodes and building a negative supply to provide bias is certainly an option. you could add a trannie, etc to get the voltage, or derive the voltage in various ways from the existing supply. if you study schematics avaliable at various sites on the web you will see some clever schemes. if you choose this route you will get back the voltage lost thru the cathode resistors and voila, more power. but...will that SOUND better?
consider this. you can also lower the plate voltage to bring things into line. low and hot, low and hot. yeah baby. you will note that the class "A" ratings in the tube manuals generaly show plate voltages well BELOW the maximum...because these tube designers were telling you something. run the tubes in the sweet spot on the curves for linearity, good sound.
look at the hammond chokes at angela.com. they got some DC resistance, hint hint.
now you have a perfectly good reason to put a choke in this supply. or even a tube rectifier. don't be afraid to cut traces and run wires to mod this thang.
you boys will learn much by studying the traces left by the ancient ones hehe. the radiotron, old schematics, and tube manuals.
[/B][/QUOTE]chromal said:
just one comment here.
bigger caps are not necessarily an "upgrade". and if this amp is "attentuated" in the low freqs the caps are NOT the reason.
it would more likely be the output trannies and measurement would be most revealing.
"soft" low end is likely the speaker/amp interface...how the amp damps your speakers. messing with the NFB can tweak this...cut and try, cut and try.
and, let me point out...the ancient ones were not very concerned with maximum power output ratings. they were happy to sacrifice the top few watts to get the first watt sounding goooood...
Re: My K-12M's doing the same thing
Nice work on the enclosure. That enclsoure will help keep the tube heat from raising the temeprature on the power transformer more than it will do so on its own. Nice tip with the USB cables.
I have not experienced the heat problems with the tubes, but the power transformer is a little on the small side and when you add the excess heat from the tubes, my power tansformer gets real hot. However, I don't think that the heat should cause problems for power tranny. Or atleast I hope not.
As Blackie noted, heat can be a problem for resistors and the power supply resistors do run hot. Upgrading to 5W resistors from 2W may be a good idea. If you use resistors with long leads, some creative bending should allow them to fit on the PCB easily.
I got this kit as cheap introduction to tube audio and let me tell you, it is like a disease (a good one though)! Since getting the kit, it has made me buy a pile of stuff .. Fostex drivers, a scope, an LC meter, a tube tester, two old tube consoles and a pile of capacitors. I now have so many project on the go I don't know where I will find the time to do them all!
Anyway, it is a fun little kit and a great introduction into tube audio.
I am planning on upgrading the capacitors soon.
Cheers,
Gio.
Powercntrl said:
Nice work on the enclosure. That enclsoure will help keep the tube heat from raising the temeprature on the power transformer more than it will do so on its own. Nice tip with the USB cables.
I have not experienced the heat problems with the tubes, but the power transformer is a little on the small side and when you add the excess heat from the tubes, my power tansformer gets real hot. However, I don't think that the heat should cause problems for power tranny. Or atleast I hope not.
As Blackie noted, heat can be a problem for resistors and the power supply resistors do run hot. Upgrading to 5W resistors from 2W may be a good idea. If you use resistors with long leads, some creative bending should allow them to fit on the PCB easily.
I got this kit as cheap introduction to tube audio and let me tell you, it is like a disease (a good one though)! Since getting the kit, it has made me buy a pile of stuff .. Fostex drivers, a scope, an LC meter, a tube tester, two old tube consoles and a pile of capacitors. I now have so many project on the go I don't know where I will find the time to do them all!
Anyway, it is a fun little kit and a great introduction into tube audio.
I am planning on upgrading the capacitors soon.
Cheers,
Gio.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.