CCS Chip blown up???
My first time here and my first tube amp (tubelab SE)...
After issues with the PSU (B+ seems to high (450V) as well as B- (290V), no power tube), I tried to get the 150V on the 5842. I have no reading (or very little, 5V or so) for both channels. Before I order new chips, I wanted to consult you folks who clearly have a better understanding of that circuit compare to me!
Could it be possible that both chips failed at the same time? One of the two seems to get very hot. I tried reading on R16 & R27 - nothing.
Any one can track me toward possible solutions???
Many thanks in advance!
My first time here and my first tube amp (tubelab SE)...
After issues with the PSU (B+ seems to high (450V) as well as B- (290V), no power tube), I tried to get the 150V on the 5842. I have no reading (or very little, 5V or so) for both channels. Before I order new chips, I wanted to consult you folks who clearly have a better understanding of that circuit compare to me!
Could it be possible that both chips failed at the same time? One of the two seems to get very hot. I tried reading on R16 & R27 - nothing.
Any one can track me toward possible solutions???
Many thanks in advance!
Last edited:
It doesn't sound like you have anything too far out of the ordinary with your voltages. I had over -200 volts on mine, and there is a recent thread about it here:
http://www.diyaudio.com/forums/tubelab/195101-tubelab-se-checkout-shows-300v-bias-voltage.html
Your B+ will also sag quite a bit when you actully have tubes in them. I assume you are setting your's up for 300b with something like a 660 volt transformer?
I had a heck of a time setting the voltage on the input tubes, going off the coupling cap and ground, I could not get a reading at all. I can't tell you why. I ended up pulling the tube about 1/3rd of the way out of the socket and using those spring clip test leads to grab the appropriate pins on the tube and setting it that way. No matter how you test it be extremely careful, those are lethal voltages
I can't say for sure that your chips aren't blown, but like you said it seems pretty unlikly that they would both go at the same time. Hope this helps.
http://www.diyaudio.com/forums/tubelab/195101-tubelab-se-checkout-shows-300v-bias-voltage.html
Your B+ will also sag quite a bit when you actully have tubes in them. I assume you are setting your's up for 300b with something like a 660 volt transformer?
I had a heck of a time setting the voltage on the input tubes, going off the coupling cap and ground, I could not get a reading at all. I can't tell you why. I ended up pulling the tube about 1/3rd of the way out of the socket and using those spring clip test leads to grab the appropriate pins on the tube and setting it that way. No matter how you test it be extremely careful, those are lethal voltages
I can't say for sure that your chips aren't blown, but like you said it seems pretty unlikly that they would both go at the same time. Hope this helps.
Thanks Jpeg. You are correct: 300B with 660V transfo. I will investigate a little more tonight based on your comments.
Your comments were helpful
Rknize,
Direct mesurement on R16/R27 (clips on both sides of each resistors, not R16/R27 to ground), shows between 0 to 5V. When playing with R12/R23, I see a voltage fluctuation (0 to 15v) for half a second or so.
The situation seems to show that there is almost no current across the input stage (5842). I see only two possibilies: bad contact like Jpeg suggested, or blown up CCS on both channel
Last edited:
Russ,
I am currently at work. I will measure tonight R10/21 but if I remember my measure from last night it was around zero.
I am aiming to get 150V at the plate as per instructions from TUBELAB. I only have around 5 volts.... Something is definitly wrong.
Alain
I'm confused. Are you saying that you have 5V at the plate or that you have 5V across R16/R27? 5V at the plate (relative to ground) is wrong. That should be 170V or whatever. 5V across the resistor is about right.
If the answer yes to both, then it sounds like you have a short across the tube. Either the tube plate is shorted to the cathode or you have a solder short under the socket.
If the answer yes to both, then it sounds like you have a short across the tube. Either the tube plate is shorted to the cathode or you have a solder short under the socket.
Russ,
I have 5V on plate to ground (fluctuating, it starts at 15v when I am powering up and falls down between 0 to 5 volts). Both tube comes from ebays - might be the cause...
I will investigate.
Many thanks Russ - I really appreciate your support.
Alain
Russ,
I tried with Raytheon and with Ampex.
I have:
B-: 297V
B+: 448V
R16: 5.75v
R17: 0V
R27: 5.79V
R28: 0V
R10: 0V
R21: 0.02V
As a note, both CCS are getting very warm with 5842 in. Does seem to happen without. I checked for possible shorts - seems OK.
I received today 6x new CCS. I will try new ones.
Alain
I thinK I figured it out!
I think I know why your voltages are reading so weird. You have a bad gound. That was the reason why I had so much trouble measuring voltages on my 5842, by chassis ground had come loose, from your readings, I bet $10 you have the same problem. Your chassis needs to be grounded back to the circuit board in addition to the 3rd prong of your outlet. I connected one of the ground wires of my RCA inputs to the chassis and now I have no trouble measuring 5842 voltage from the coupling cap and ground.
I hope this helps.
I think I know why your voltages are reading so weird. You have a bad gound. That was the reason why I had so much trouble measuring voltages on my 5842, by chassis ground had come loose, from your readings, I bet $10 you have the same problem. Your chassis needs to be grounded back to the circuit board in addition to the 3rd prong of your outlet. I connected one of the ground wires of my RCA inputs to the chassis and now I have no trouble measuring 5842 voltage from the coupling cap and ground.
I hope this helps.
B+ and B-: look fine.
R16/R27: if that is the voltage *across* that resistor, it looks fine. The CCS are doing their job. They will get fairly toasty after a while: 448V - 170V = 278V * 12mA = 3.3W. If there is a short, then they have all of B+ across them and will get very hot indeed. You have the CCS heatsinks insulated, right?
R17/R28: these are just gate stoppers for the CCS. Practically no current runs through it. 0V is expected
R10/R21: The voltage *across* this resistor should be around 6V or so.
Jpeg,
I have not grounded the board to ground outlet - I waited to do that when I would do the final assembly with the chassis. Could it be that simple??? Man, I would feel idiot if it is the case, knowing how reference grounding is important!!!
I will try tonight. Thanks you are helpful.
Russ,
Do you mean the CSS needs to be insulated from the heat sink? No I did not - I have insulated all the other but not the two CCS, trusting pictures of the Tubelab manual. Do you think it is critical since both (Anode) are connected to B+? Or am I missing something?
Thanks for the helpful info. I really appreciate your support. I own you one!
Alain
Sorry, what I meant was that they need to be insulated from the chassis. If you are using the heat sinks from the parts list and they are not mounted to the chassis at all, then you don't need to insulate them (note that the heatsink will have the full B+ voltage on it).
You don't need an earth ground on the chassis for the amp to work (though it is needed for safety reasons). The PCB has it's own ground that it uses because the amp and power supply are all together. Since your B+/B- measurements make sense, I assume you are using the PCB ground as a reference for your meter.
Just to clarify this one more time: when you place both meter probes on each side of R16/27, you get ~5V. When you ground the meter to the PCB power supply ground and measure both sides of the resistors with the meter, you get 0V on the tube side and 5V on the CCS side? What do you get with no tube in place?
You don't need an earth ground on the chassis for the amp to work (though it is needed for safety reasons). The PCB has it's own ground that it uses because the amp and power supply are all together. Since your B+/B- measurements make sense, I assume you are using the PCB ground as a reference for your meter.
Just to clarify this one more time: when you place both meter probes on each side of R16/27, you get ~5V. When you ground the meter to the PCB power supply ground and measure both sides of the resistors with the meter, you get 0V on the tube side and 5V on the CCS side? What do you get with no tube in place?
Ok, here it is:
Without tube (still heat from heatsinks):
Ground to:
R16 tube: 17V, R16 IC: 449V, both sides: 0V
R17 tube: 17V, R17 IC: 17V, both sides: 0V
R27 tube: 17V, R27 IC: 449V, both sides: 0V
R28 tube: 17V, R28 IC: 17V, both sides: 0V
R10 & 32 both sides: 0V
With tubes on (heat from heatsinks):
Ground to:
R16 tube: 6V, R16 IC: 13V, both sides: 6.5V
R17 tube: 6V, R17 IC: 6V, both sides: 0V
R27 tube: 3V, R27 IC: 10V, both sides: 5V
R28 tube: 3V, R28 IC: 3V, both sides: 0V
R10 & 32 both sides: 0V
Ground to tube: 6.6 & 4.25V
I tried for potential poor coupling cap (I rised one leg for both coupling cap) = no change in the data.:
I checked again for potential shorts - nada.
I also grounded the board to the grid - nada.
However, both tube sockets are hooked up through 4 inches of wire (not directly welded to the board). Am I facing oscillations or some weird phenomena?
My next step is to change both ICs. I don`t have a good feeling for the end result but I need to give it a try...
Alain
It does not need to be grounded to work, but, if you are testing voltage values against the ground, it does need to be gounded. For safety, it should be grounded to the 3rd prong of an outlet, but the ground I am refering to is the return path back to the power supply. (It is kind of confusing) If the black lead of your meter is not connected to a to the return path, you may be 0 readings when if fact there is power. If your chassis is not connected to the board ground somewhere, you should make sure you are connecting your test lead to a good ground connection on the board such as the hole labled GND 2 on the board, near in between C5 and where your inputs hook up.
When you tested your plate votage on your 5842s and you put the red lead of your tester on the coupling capacitor, where did you put your black lead?
When you tested your plate votage on your 5842s and you put the red lead of your tester on the coupling capacitor, where did you put your black lead?
JPeg,
I connected the probe to Gnd2 (bottom of the board) - Always did since the begining. The board is on my work bench, not yet in a box
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.