1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

Millett "Starving Student" hybrid amp

First
 
Back
289 290 291 292 293 294 295 296 297 298
300 301 302 303 304 305 306 307 308 309
Next
 
Last
  1. revolink24
    Yes! I just managed to score 5 19J6s for $8 a pop on ebay. Should I keep them just to have (lots) of spares, sell them, or build another amp? Building another amp sounds redundant albeit fun. I could always build and sell one.
     
  2. Juaquin
    Do whatever your heart desires, but if you do decide to sell them be sure to let people know here - I'm sure many are looking for them.
     
  3. th3bl0b
    Oh, I thought about pictures, but the insides of it looks like spaghetti.

    After a while, I realized that I had my rocker switch incorrectly wired. Nothing was getting through. There was no voltage between the ground and anything beyond the rocker, but once I took out the rocker. The tubes started to light up. Boy, that felt great when they did. I realized that all my work wasn't for anything!

    I'd put pics up, but my enclosure needs some work. I plan to try again as far as making an enclosure goes.

    Thanks for your help! Let me know when the kits are ready cause I'll definitely have to jump on that.

    Quote:

    Originally Posted by tomb /img/forum/go_quote.gif
    Yes, the tubes are supposed to lightup. Better post some pics so we can take a look at what's going on.

    Are you able to measure whether there's power coming from the walwart when you turn it on? The Cisco power supply has a very good circuit protector. If you have a short in your amp, it will simply refuse to supply voltage. Measured separately and disconnected, it will appear to function just fine under those circumstances. Try and measure from the wires coming from your power socket on the amp and see if it's supplying voltage.




     
  4. tomb
    Quote:

    Originally Posted by th3bl0b /img/forum/go_quote.gif
    Oh, I thought about pictures, but the insides of it looks like spaghetti.

    After a while, I realized that I had my rocker switch incorrectly wired. Nothing was getting through. There was no voltage between the ground and anything beyond the rocker, but once I took out the rocker. The tubes started to light up. Boy, that felt great when they did. I realized that all my work wasn't for anything!

    I'd put pics up, but my enclosure needs some work. I plan to try again as far as making an enclosure goes.

    Thanks for your help! Let me know when the kits are ready cause I'll definitely have to jump on that.




    Glad to hear you found the problem!
     
  5. Entropy1
    I was reading earlier in this or another thread about removing coupling caps on the source when going into an amp. It's possible I misunderstood this, but is this appropriate? Could I test the DC by having no audio playing, and measuring this directly? If this does turn out to be at least somewhat accurate, what would be a reasonable level that it should be under to do this? Would it be good to change the input pot or anything if I do this? Thanks for the help.
     
  6. tomb
    Quote:

    Originally Posted by Entropy1 /img/forum/go_quote.gif
    I was reading earlier in this or another thread about removing coupling caps on the source when going into an amp. It's possible I misunderstood this, but is this appropriate? Could I test the DC by having no audio playing, and measuring this directly? If this does turn out to be at least somewhat accurate, what would be a reasonable level that it should be under to do this? Would it be good to change the input pot or anything if I do this? Thanks for the help.



    You may be stating this in the reverse. Input caps on an amp may be deleted if the source is found to have no DC offset. However, the Starving Student uses NO input capacitors.

    As for the source, if the source is using output caps, there is normally a reason - there's too much votage to subject to an amp or a load. There would be no reason to apply capacitors to the output of a source unless there is significant DC present. That certainly doesn't mean that you can remove them, though.

    I have heard of isolated instances where people used output caps just to "flavor" the sound, even if they weren't needed to block DC - but that's usually someone's idea of a mod and the evidence is obvious.
     
  7. amc
    Quote:

    Originally Posted by tomb /img/forum/go_quote.gif
    ...it looks like maybe R13 is not soldered on one end - that's the resistor up against the power input terminal block. That resistor divides the voltage supply between the two tubes and may cause some of your symptoms if it's not soldered.



    Thanks tomb. I checked underneath and it does in fact have a solder volcano on the pin but it did not suck up through the hole. I must have been impatient and not heated the pad enough. I managed to get my iron between the cap and terminal block and apply some solder to the pad and lead on the top side. Hopefully thats got it if it was shorting or cold.

    Attachment 24743

    Quote:

    Originally Posted by tomb /img/forum/go_quote.gif
    A second suspicion is your tube LEDs....If R13 or any other resistor doesn't check out (check them all - for value, too), you might try de-soldering those leads...Take a look at those things and let us know.[​IMG]




    I checked all the resistors for suspect joints again - everything else looks good with good wicking from the bottom. I checked the resistance of all of them. Here is what I measured:

    R1 33.07K
    R2 110.0K *
    R3 2K
    R4 120.0K *
    R5 2K
    R6 1.998K
    R7 33.01K
    R8 105K *
    R9 2K
    R10 108K *
    R11 2K
    R12 2K
    R13 2K
    R14 0.997K
    R15 0.998K
    R16 99.8K
    R17 99.9K

    Notice the ones with the Astrix (*). For the life of me I can not get an accurate reading of these 4 resistors. As soon as I put the probes down the meter drifts upwards from about 110 at a constant rate - never zeroing in on the expected value of 220K (markings on the resistors are 2203 so thats 220K right?) Its painful. I have left the probes down for close to 10 minutes and only reached 140 or so ohms. any explanation of this behavior?


    Quote:

    Originally Posted by the_equalizer /img/forum/go_quote.gif
    Great pics, thanks! As TomB says, check R13, one of the pads does look as if it weren't soldered. If that doesn't solve it... what's the voltage across R4 (and across R10) ?



    So already Im weirded by R4 and R10 because of the resistance. Measuring B+ to one side of the resistor measures 48.1v and B+ to the other side of the resistor measures 47.6v consistent for both R4 and R10. So there is a 0.5v drop. Measuring with the probes directly across the leads of each resistor is a bit different. After measuring from B+ to the resistor leads I try and measure between the leads. I get about 4.5mV and it drops consistently so that after a minute I am down to 1.9mV

    So still stuck hopefully this helps in narrowing down the issue... fingers crossed.
    vbattach24743.jpg
     
  8. tomb
    Quote:

    Originally Posted by amc /img/forum/go_quote.gif
    Thanks tomb. I checked underneath and it does in fact have a solder volcano on the pin but it did not suck up through the hole. I must have been impatient and not heated the pad enough. I managed to get my iron between the cap and terminal block and apply some solder to the pad and lead on the top side. Hopefully thats got it if it was shorting or cold.

    Attachment 24743




    I checked all the resistors for suspect joints again - everything else looks good with good wicking from the bottom. I checked the resistance of all of them. Here is what I measured:

    R1 33.07K
    R2 110.0K *
    R3 2K
    R4 120.0K *
    R5 2K
    R6 1.998K
    R7 33.01K
    R8 105K *
    R9 2K
    R10 108K *
    R11 2K
    R12 2K
    R13 2K
    R14 0.997K
    R15 0.998K
    R16 99.8K
    R17 99.9K

    Notice the ones with the Astrix (*). For the life of me I can not get an accurate reading of these 4 resistors. As soon as I put the probes down the meter drifts upwards from about 110 at a constant rate - never zeroing in on the expected value of 220K (markings on the resistors are 2203 so thats 220K right?) Its painful. I have left the probes down for close to 10 minutes and only reached 140 or so ohms. any explanation of this behavior?




    So already Im weirded by R4 and R10 because of the resistance. Measuring B+ to one side of the resistor measures 48.1v and B+ to the other side of the resistor measures 47.6v consistent for both R4 and R10. So there is a 0.5v drop. Measuring with the probes directly across the leads of each resistor is a bit different. After measuring from B+ to the resistor leads I try and measure between the leads. I get about 4.5mV and it drops consistently so that after a minute I am down to 1.9mV

    So still stuck hopefully this helps in narrowing down the issue... fingers crossed.




    You can't measure resistances on a populated PCB. Too many of them are in parallel with other resistors or other items in the circuit, including capacitors that the DMM will charge and discharge as you found out.[​IMG][​IMG] Unfortunately, it can be a meaningless exercise.

    It's best always to read voltages across the resistors on a populated board - or measure voltage at points with one probe at ground. Go back and try again.[​IMG]

    EDIT: Maybe the equalizer can make something of your voltage readings. The ~47.6 - 48V is close enough for my tastes, but it sounds like you're reading across a capacitor again on the other one.

    EDIT #2: I just looked at your pic and while dark, it appears that you may have torqued down the MOSFETs so much that it cut into the insulators. Going against what I just said about measuring resistances, with the amp OFF, try seeing if you get a resistance across the MOSFET tab and an unanodized part of the case. If you don't read infinite resistance or something similar, you may have shorted out the MOSFETs.
     
  9. rauschie
    Hello guys, I'm quite certain that I'm not the first one asking this, but I don't have time to read 300 pages, so If there is already an answer in the forum, please just link it then ignore me, but now let me have this absolutely lame question:
    I'm planning to build this amp, I already have the tubes and the mosfets (which are the most difficult parts to find in Hungary) so there's not much left to begin.
    The problem:
    At the moment, I'm using Senn HD555's, but as fast as I can afford, i'm going to change to 580's, 600's or 650's, so I want the amp to perform well both with the 300 ohm headphones and as well as the sensitive 555's, therefore I'm supposed to build in a gain-switch.
    The question itself:
    Which components do I have to replace, and what type of components do I have to use?
    As far as I could understand, I'm going to have to change some resistors or capacitors, or rather have two different type of them with a switc with which I can choose which to use.
    I'm quite a newbye at DIY electronics, so please have mercy on me and try to answer in an easy-to-understand way.
    Sorry for my bad english and Thanks for the answer!
     
  10. Juaquin
    Instead of a gain switch, it would probably be easier to switch the input resistors. Have the switch short across them for a "higher" gain (technically it's not a higher gain, just not reducing the input signal) and then switch over to the 100k resistors for "less" gain (because the resistors are reducing the input signal).
     
  11. sourced
    Quote:

    Originally Posted by rauschie /img/forum/go_quote.gif
    ... gain-switch.
    The question itself:
    Which components do I have to replace, and what type of components do I have to use?
    As far as I could understand, I'm going to have to change some resistors or capacitors, or rather have two different type of them with a switc with which I can choose which to use.
    I'm quite a newbye at DIY electronics, so please have mercy on me and try to answer in an easy-to-understand way.
    Sorry for my bad english and Thanks for the answer!




    This I found in a similar thread (so not in the 300 pages, but still on head-fi).

    http://www.head-fi.org/forums/f6/mil...-build-332593/

    Scroll past the pictures until you see a hand-drawn diagram. That isn't really a 'gain' switch per se, but it will have a similar effect. Is that what you were trying to figure out?
     
  12. RedLeader
    Quote:

    Originally Posted by revolink24 /img/forum/go_quote.gif
    Yes! I just managed to score 5 19J6s for $8 a pop on ebay. Should I keep them just to have (lots) of spares, sell them, or build another amp? Building another amp sounds redundant albeit fun. I could always build and sell one.



    Some of us want some....
     
  13. the_equalizer
    Quote:

    Originally Posted by amc /img/forum/go_quote.gif
    Thanks tomb. I checked underneath and it does in fact have a solder volcano on the pin but it did not suck up through the hole. I must have been impatient and not heated the pad enough. I managed to get my iron between the cap and terminal block and apply some solder to the pad and lead on the top side. Hopefully thats got it if it was shorting or cold.

    I checked all the resistors for suspect joints again - everything else looks good with good wicking from the bottom. I checked the resistance of all of them. Here is what I measured:

    R1 33.07K
    R2 110.0K *
    R3 2K
    R4 120.0K *
    R5 2K
    R6 1.998K
    R7 33.01K
    R8 105K *
    R9 2K
    R10 108K *
    R11 2K
    R12 2K
    R13 2K
    R14 0.997K
    R15 0.998K
    R16 99.8K
    R17 99.9K

    Notice the ones with the Astrix (*). For the life of me I can not get an accurate reading of these 4 resistors. As soon as I put the probes down the meter drifts upwards from about 110 at a constant rate - never zeroing in on the expected value of 220K (markings on the resistors are 2203 so thats 220K right?) Its painful. I have left the probes down for close to 10 minutes and only reached 140 or so ohms. any explanation of this behavior?




    As TomB has said, you can't measure a resistor's resistance value while it's connected to a circuit. Other resistors in the circuit, capacitors, semiconductors, etc. will form a current path and you'll get all kinds of crazy readings.


    Quote:

    Originally Posted by amc /img/forum/go_quote.gif
    So already Im weirded by R4 and R10 because of the resistance. Measuring B+ to one side of the resistor measures 48.1v and B+ to the other side of the resistor measures 47.6v consistent for both R4 and R10. So there is a 0.5v drop. Measuring with the probes directly across the leads of each resistor is a bit different. After measuring from B+ to the resistor leads I try and measure between the leads. I get about 4.5mV and it drops consistently so that after a minute I am down to 1.9mV

    So still stuck hopefully this helps in narrowing down the issue... fingers crossed.




    The reading of 4.5mV across R4 and R10 shows there's something seriously amiss. The reading should be half the power supply voltage (between 23 and 24 V). So there must be a big short circuit somewhere.

    As TomB says, check the MOSFETs for shorts to the case (did you install the plastic shoulder washer between the nut and the MOSFET tab? )

    cheers!

    cheers!
     
  14. amc
    Quote:

    Originally Posted by the_equalizer
    As TomB says, check the MOSFETs for shorts to the case (did you install the plastic shoulder washer between the nut and the MOSFET tab? )



    I did install the washers but i really torqued down the heatsinks before reading about damaging the burquist pads.

    Quote:

    Originally Posted by tomb /img/forum/go_quote.gif
    you may have torqued down the MOSFETs so much that it cut into the insulators. Going against what I just said about measuring resistances, with the amp OFF, try seeing if you get a resistance across the MOSFET tab and an unanodized part of the case. If you don't read infinite resistance or something similar, you may have shorted out the MOSFETs.



    Indeed you are right. I read resistance between a scratch i put in the underside of the lid and both mosfet tabs. The mosfets will still be OK right? I just need new insulators?

    I got some at a local electronics store but they are not the same as the burquist pads you supplied but mica sheets. Can I use those? Should I clean off the buquist pads or just leave them in place...

    Will work on it tonight and keep you guys posted. Thanks for your help so far.
     
  15. simwells
    Quote:

    Originally Posted by tomb /img/forum/go_quote.gif
    An RK097 would probably be a realistic option, but most are only available in 10K. You'd have to re-work your input resistor sizing accordingly. If you've used the PCB, the pin spacing is not going to match (there'd be more for the switch, anyway).



    Wasn't planning on going the PCB route, but my lack of electronics means I think I'll stick with a normal switch! No idea how to recalculate other values.
     
First
 
Back
289 290 291 292 293 294 295 296 297 298
300 301 302 303 304 305 306 307 308 309
Next
 
Last

Share This Page