I built a stereo single ended kt88 amp based on the diyaudioprojects michael abdellah design... i fired it up last night for the first time, and it ran great all evening... I ran it a again for a few hours this afternoon and then after running for about 1 hour this evening the fuse blew out......
afterwards, I flipped the amp over to look for any signs of a short or anything else out of the ordinary and I wondered what my first course of action should be? is this a rectifier problem? a tube problem? I am just not sure exactly where to start hunting ....
afterwards, I flipped the amp over to look for any signs of a short or anything else out of the ordinary and I wondered what my first course of action should be? is this a rectifier problem? a tube problem? I am just not sure exactly where to start hunting ....
You haven't said what size and type of fuse you're using. (slow, fast & rating) Could it be undersized? Can I assume it's in the power transformer primary? Where did you source your KT88's? Were they properly burned-in by a dealer, or are they raw stock? Todays output tubes are a sorry lot to say the least, especially if they are Chinese made. Unless they were burned-in and aged for stability, this could well be your problem.
To correctly trouble shoot this condition you will need to either try "known to be stable" KT88's, or you must monitor cathode current of both tubes separately (across a 10 ohm resistor) and watch for upward current creep over time.
If you're using a Chinese rectifier tube, that's another suspect. But less so then the outputs.
To correctly trouble shoot this condition you will need to either try "known to be stable" KT88's, or you must monitor cathode current of both tubes separately (across a 10 ohm resistor) and watch for upward current creep over time.
If you're using a Chinese rectifier tube, that's another suspect. But less so then the outputs.
thanks for the advice from both of you. I am at work now but will try to get more information back on here tonight. I can show you the schematic and list all the tube types I am using ... as far as taking measurements I'm not sure exactly how to measure each stage ...
For now, I can tell you what I do know. The power tubes i'm using are Electro-Harmonix kt88s and they were purchased from tubedepot... idk if they burn them in there but they are a match set. The rectifier is a Sovtek 5U4G. The fuse is a slow blow but i'm not sure on the amp rating.
like i said I will get on here tonight and give you all the info i can. I put in a new fuse and played it for another hour last night... the new fuse hasn't blown out yet.
For now, I can tell you what I do know. The power tubes i'm using are Electro-Harmonix kt88s and they were purchased from tubedepot... idk if they burn them in there but they are a match set. The rectifier is a Sovtek 5U4G. The fuse is a slow blow but i'm not sure on the amp rating.
like i said I will get on here tonight and give you all the info i can. I put in a new fuse and played it for another hour last night... the new fuse hasn't blown out yet.
I have built an amp like porkchops (his schematic is here) and I followed it almost exactly....
Rectifier : 5u4g sovtek
Power tubes: kt88 Electro Harmonix
preamp tube: 6n1p sovtek
My power transformer is a hammond 273bx and the power supply output is sitting at 421 Volts when the amp is running... I just checked it a minute ago... the schematic calls for 400 volts... is it being that high is the problem??
I am using a 1amp 250v slow blow fuse and I tried to calculate the draw from the power supply this morning (I know very little about this) and I came up with .9 amps... my next question is whether or not I should just bump up to a 1.5 amp fuse or something???
I have run the amp another hour today and the fuse didn't blow out... am I in danger of hurting the amp by running it right now until I get the fuse issue solved??? or should I not stress about it???
Last question what would happen if I changed the 560 ohm resistors from pin 8 to ground on the power tubes to a 470 ohm resistor... I never plan on running el34s in this amp.
thanks in advance to any and all help!
Rectifier : 5u4g sovtek
Power tubes: kt88 Electro Harmonix
preamp tube: 6n1p sovtek
My power transformer is a hammond 273bx and the power supply output is sitting at 421 Volts when the amp is running... I just checked it a minute ago... the schematic calls for 400 volts... is it being that high is the problem??
I am using a 1amp 250v slow blow fuse and I tried to calculate the draw from the power supply this morning (I know very little about this) and I came up with .9 amps... my next question is whether or not I should just bump up to a 1.5 amp fuse or something???
I have run the amp another hour today and the fuse didn't blow out... am I in danger of hurting the amp by running it right now until I get the fuse issue solved??? or should I not stress about it???
Last question what would happen if I changed the 560 ohm resistors from pin 8 to ground on the power tubes to a 470 ohm resistor... I never plan on running el34s in this amp.
thanks in advance to any and all help!
If you update your location information we would know whether you were in a 120V or 230V country - and that would help us to determine what the right primary fuse would be.
If you are in 120V land you probably need a 2A - 2.5A slow blow to account for things like magnetic inrush and thermally fatiguing the fuse over many power cycles, if in 230V land then something like 1.5A slow blow might be right.
The 420V supply is reasonably in tolerance when you take into account possible variations in line voltage, and the operating point of your amp.
Note that if you intend to use just KT88 then a 470 resistor is fine, the increased current will result in a slight reduction of B+ as well.. (Addressing your concern above.) Also important is the fact that in metal based KT88 pin 1 is connected to the external metal shell if present - tying this to pin 8 will result in the moderately hazardous cathode voltage appearing there - so ground pin 1 to chassis star ground and pin 8 goes to your cathode bias resistor and cap..
If you indeed built it like porkchop did then you need to connect the "0" ohm tap on each OPT to your star ground for safety reasons.
If you are in 120V land you probably need a 2A - 2.5A slow blow to account for things like magnetic inrush and thermally fatiguing the fuse over many power cycles, if in 230V land then something like 1.5A slow blow might be right.
The 420V supply is reasonably in tolerance when you take into account possible variations in line voltage, and the operating point of your amp.
Note that if you intend to use just KT88 then a 470 resistor is fine, the increased current will result in a slight reduction of B+ as well.. (Addressing your concern above.) Also important is the fact that in metal based KT88 pin 1 is connected to the external metal shell if present - tying this to pin 8 will result in the moderately hazardous cathode voltage appearing there - so ground pin 1 to chassis star ground and pin 8 goes to your cathode bias resistor and cap..
If you indeed built it like porkchop did then you need to connect the "0" ohm tap on each OPT to your star ground for safety reasons.
Your amp is likely running around 8 or 9 mA on the 6n1p - heck..lets call it 10...and with a 560 on the KT88 probably around 75mA each tube. So lets call that 160mA on the high voltage...
On the heaters side...you are running 600mA on the 6n1p and 1.6A on each KT88.
On the rectifier side the 5u4g pulls 3A and your xformer is rated at 3A...I might start there and try a 5ar4 or maybe temporarily solder in a couple 1n4007's and see if that alleviates the problem.
I have a Tubelab SimpleSE and it has a 1.5A slow blow and has many hours on it. I am running a 12AT7 and a pair of KT90's running at about 89mA typically on SS rectification. If I need to kick the B+ down a little I throw in a 5ar4 with no issues
On the heaters side...you are running 600mA on the 6n1p and 1.6A on each KT88.
On the rectifier side the 5u4g pulls 3A and your xformer is rated at 3A...I might start there and try a 5ar4 or maybe temporarily solder in a couple 1n4007's and see if that alleviates the problem.
I have a Tubelab SimpleSE and it has a 1.5A slow blow and has many hours on it. I am running a 12AT7 and a pair of KT90's running at about 89mA typically on SS rectification. If I need to kick the B+ down a little I throw in a 5ar4 with no issues
"If you are in 120V land you probably need a 2A - 2.5A slow blow to account for things like magnetic inrush and thermally fatiguing the fuse over many power cycles, if in 230V land then something like 1.5A slow blow might be right."
i've built the same amp, had the same prob till i change the fuse from .5a to 2.5a no more blowing fuse.
the amp been in my main system for about 2 years now with no issues.
i've built the same amp, had the same prob till i change the fuse from .5a to 2.5a no more blowing fuse.
the amp been in my main system for about 2 years now with no issues.
I think it is best to use as low rating a fuse as possible, to have proper protection, and have a simple timed relay to limit the inrush. If you don't have extra winding in your power transformer maybe this will work
A NE555 timer can be used to drive the relay to short out a 22 Ohm resistor after 4 seconds.
(P/T secondary 6.3 AC)---->7 V DC -----> Timed relay ckt
|
|
--->(6.3 AC to heaters)
A NE555 timer can be used to drive the relay to short out a 22 Ohm resistor after 4 seconds.
(P/T secondary 6.3 AC)---->7 V DC -----> Timed relay ckt
|
|
--->(6.3 AC to heaters)
costis... i appreciate the advice and while yours is probably the best option I think for now at least I will bump to a 2amp slow blo fuse.... it's a quick and easy solution.
I am located in OHIO and am 120v sorry for neglecting to mention that earlier... I am going to try going down to that 470ohm resistor like i mentioned and also I will try to follow kevin's advice.
I am still a little confused though. I do not currently have the output trannies grounded from 0 to ground so i can go ahead and do that... but as far as the pins on the kt88... should i untie pins one and 8 and ground like you said? Did porkchop just have those tied in the original schematic because he was trying to run el34s as well?
last question... to open a can of worms... i was looking at some of the mods kegger did to this amp design... he changed a lot of the resistor values and tied a few things differently... but the big thing I am interested in is his tying the bottom of the input tube to ground with leds instead of a resistor/cap combo....
If I were to take what I have now on the schematic I am working from is that something I could try out?
I am located in OHIO and am 120v sorry for neglecting to mention that earlier... I am going to try going down to that 470ohm resistor like i mentioned and also I will try to follow kevin's advice.
I am still a little confused though. I do not currently have the output trannies grounded from 0 to ground so i can go ahead and do that... but as far as the pins on the kt88... should i untie pins one and 8 and ground like you said? Did porkchop just have those tied in the original schematic because he was trying to run el34s as well?
last question... to open a can of worms... i was looking at some of the mods kegger did to this amp design... he changed a lot of the resistor values and tied a few things differently... but the big thing I am interested in is his tying the bottom of the input tube to ground with leds instead of a resistor/cap combo....
If I were to take what I have now on the schematic I am working from is that something I could try out?
I estimate your amp uses about 1.2A (estimated 105 W, 95% efficiency, 80% power factor). Decreasing the cathode resistor will increase that a bit more...
Transformer secondary should be grounded for safety, could also help stability in some cases by acting as a shield (only likely to be an issue in high gain guitar amps). Also must be grounded if you want to use global negative feedback.
Pin 1 connection can be removed if you don't intend to use EL34, but there's no reason to remove it unless you use metal base 6550/KT88 with shell connected to pin 1.
The LED provides "fixed" bias (constant voltage) due to its low dynamic resistance and eliminates the need for a cathode bypass.
Transformer secondary should be grounded for safety, could also help stability in some cases by acting as a shield (only likely to be an issue in high gain guitar amps). Also must be grounded if you want to use global negative feedback.
Pin 1 connection can be removed if you don't intend to use EL34, but there's no reason to remove it unless you use metal base 6550/KT88 with shell connected to pin 1.
The LED provides "fixed" bias (constant voltage) due to its low dynamic resistance and eliminates the need for a cathode bypass.
I'd double the size of the fuse. Use a 2A fuse. You can't damage anything by putting in a to large fuse. The purpose of the fuse is to limit the damage to the amp if something fails. For example if a diode shorts. So you want a fuse small so that it will do a good job of containing any damage after a part fails but not so small that you get "nuisance" fuse blowing. Likey with your 1A fuse there is a slight voltage surge on the AC line and that kills the fuse. Going up to 2A Slow Blow is a good compromise.
About changing the cathode resistor. I like to use pots in amps so I can adjust resistance. I think it is best to use a few resistors in series, First a 1 ohm, simply to make measuring current easy, then a fixed resistor then a 100 ohm pot. Most of the voltage drop is in the pot which takes the heat off the pot. Then I can put a volt meter across the 1R resistor and read amps directly with no math involved (other than divide by 1 which is easy) Turn the pot to get the current I want. Every tube is different even if they are "matched" it is never perfect. Do make sure to use a pot rated for 5W or more and choose the largest value fixed resistor you can. So I gues I say to got even further then simply changing the resistor - make it variable then set it every time you swap a tube
Tom Bavis and tim614 pretty much summed up what you need to know.
1.5 to 2 amp fuse would be fine and proved ample protection.
Decreasing the cathode resistor will only cause the output tubes to draw more current. Class A is pretty close to 50%, No need to draw more current than you need.
Regards, Ron
1.5 to 2 amp fuse would be fine and proved ample protection.
Decreasing the cathode resistor will only cause the output tubes to draw more current. Class A is pretty close to 50%, No need to draw more current than you need.
Regards, Ron
That is true, but the hard part is in the last four words knowing exactly what you need. I'd think you'd want the no signal quiescent current to be on the f latest part of the tube's characteristic curve. Distortion is lowest at that point. Mostly we simply let the designer guess where this point is. it gets you pretty close. But it will be slightly different for every tube. Tubes are 20% tolerance devices at best.
That said, given a choise I'd use the largest value of resistor that gives good result. This means the current is the minimum that gives good result and that the heat produced is minimum. The parts should last longer if they run cooler.
Awhile back I bought a distortion meter. It measures THD directly and it is a good way to set tube biases. Meters are cheap on eBay. But just as good, maybe better are spectrum analyzer (FFT) software. Some is available for free. These work so well you may not want to look because you can see all the "trash" your amp is creating from a clean input signal. You can set bias so as to minimize the "trash" (as long as you stay within the tube's limits)
I guess what I'm saying is that yes you can change the resistor but it's pointless to do so unless you have some method to measure the result of the change. Like I said, it's those last four words how to know what current it is you need.
Thanks so much for all your responses guys...... its awesome to have a network of people so willing to help out. I won't bump this thread for a week or so until I make some changes and do some listening tests.
Ron ... I assume you still remember me ... I visited your "museum" a few years ago.... I'll have to pm you sometime! Hope you are well.
thanks again
7
Ron ... I assume you still remember me ... I visited your "museum" a few years ago.... I'll have to pm you sometime! Hope you are well.
thanks again
7
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.