MEDIC ! My Carvin just crapped out on me

[Nuuska]

140VAC @ C25-D3 is ok - against ground.
80VAC @ D3-R33 is ok, too.

But that positive voltage @P7 etc indicates capacitors C22 and/or C23 to be leaking DC - thus raising the bias-voltage to be haywire.

Get two new ones and replace both - since you have positive voltages @ pin 2 on V4 & V5

Capacitors are cheap - let's hope the output x-former still works. You can measure the ohm-reading between RED-BLU & RED-BRN - they should be same - within 5-10% - pull the connector & measure then.

 

[JohnW63]

Like some of the measurements I've taken, the voltage starts higher , around 6 volts, and as it warms up they continue to drop. They settle out at +0.5 vdc on V4 and +1.14 vdc on V5.

Shouldn't D3 be only allowing the negative half of the wave form, with it pointed that way ?

 

[JohnW63]

I checked out the output transformer already. I don't recall the values, but I can check again.

86 and 95 ohms.

 

[JohnW63]

I ordered a pair of the caps. I'll post when they are installed.

 

[JohnW63]

Installed the new caps. It seems to be an amp again.

R31 side of P7 = -13.3v
P32 Side of P7 = -16.4v
Center of P7 = -14.8v

Pins 2 of the tubes are the same as the center of P7. -14.8v

By having not quite the -17v as per the diagram, am I losing some headroom, since the grid isn't as strong ? Are these voltages to be expected with resistors that are probably +- 10% values ?

 

[davismanLV]

So wait..... most of this is jibberish to me but..... is it fixed do you think?? I've followed the whole thing!!

 

[JohnW63]

It is acting like and amp again. I plugged into it and twisted knobs. Guitar stuff came out. Very big step in the right direction. ( Jibberish ahead alert ) Being picky and wanting things to be right up to spec, I wonder if I can change a resistor before it gets to the bias adjustment potentiometer to get more volts. The schematic says -17 volts, and I'm running below -15v. If my reading and old electronics memory is correct, that means there is more current going from cathode to anode and the grid is not suppressing it as much as it should. Since there isn't -17 volts going to the bias adjustment pot, I can't turn the dial and make it -17. I'll stare the the diagram some more and see where I might be able to dial up more negative juice.

Tom,

It's almost really fixed. It could even BE fixed, if the more experienced amp guys say that sort of difference is within "acceptable" range of -17 volts to the power tube grids. But.... I really like things that are really close to the hoped for spec.

 

[davismanLV]

Thanks, John. It's been fun reading along with this thread and although it's WAY over my head, I still find it interesting. You're saying you just want it a tad more exact than ballpark, is that right? Which I DO understand...... keep on keeping on!!

 

[Nuuska]

Hello

Good to hear of success. You could swap resistors R31 & R32 - that would shift the range of trim pot P7 towards more negative.

 

[GAD]

Do wee need a Huzzah? I think we need a Huzzah!

 

[JohnW63]

The bias pot is close to middle, so, couldn't I dial it to have more resistance on the far side and less on the incoming side ? I was going to do the math and see if that was going to be enough of a swing to get -17v.

 

[DThomasC]

The reason that the bias pot is there is to allow adjustment of the idle current through the tubes. Tubes vary. Different tubes will require a different voltage on the grid. That's why the adjustment pot is there.

It says on the schematic "SET BIAS FOR: 2-EL84S = 40MA." Evidently, the designer of the amp thinks that around 40mA is about right. That's more important than the -17V that's written on the schematic. That was there just for troubleshooting purposes.

Anyway, there are different ways to measure idle current, but Carvin built in a way to measure it directly. Written on the schematic near the bias pot is:

BEFORE ADJUSTING
MOVE QC2 TO QC9
TEST ACROSS QC2 AND QC9

A section snipped from the schematic shows this:

If you look at the schematic, you'll see that the wire that's connected to QC2 comes from the power supply, so it has something like 350VDC on it and it feeds the EL84 plates through the output transformer. Evidently, Carvin recommends moving the wire from QC2 to QC9 which goes nowhere. When you connect your DMM between QC2 and QC9 all of the current will go through the meter. Then you adjust the pot until it reads something close to 40mA. It doesn't need to be exact.

Edit: I should add that your meter probes should be securely connected before you turn on the amp and stay connected until you turn the amp off. If the connection comes and goes the output transformer could be damaged from what the old TV guys called flyback.

 

[JohnW63]

I saw that instruction , written on the board. I was going to try and figure out where to take that measurement from. Now I know.

 

[JohnW63]

Finally got around to setting the bias. Moved the line to QC2 to QC9. Hooked the leads from the meter to them and set the meter to 200mA on one meter, and left the other one in auto range. I got zip on both. Yes, I moved the positive leads to the Amp socket on the meters. I'll check both meters on a known source to make sure the internal fuse isn't blown.

 

[JohnW63]

Finally got some fuses in for the multi-meter. The first set got eaten by the USPS partway here. At center, the bias current in 23mA. At max, it's 33mA. So, I can't get it to the 40mA setting they would like on the board. I either have not enough voltage off the full wave rectifier or some thing else keeping the current low. I'll go back through the measurements on R36-34 and see if that is where the issue might be.

 

[JohnW63]

Voltage checks after replacing the filter caps.

At the D6-D4 junction, I have 370vdc. Much closer to the spec of 360. So, the voltage supplied here is not the problem.
I measured the voltage across some resistors to calculate the current. At the big 10 watt resistor, R36 I have 2.0Vdc which means I have 5.7mA going that direction and the bias current left over from the current out of the rectifier going to the output transformer. The R36..R35...R34 leg ends up going to the output ends of Tube 3. I need to think about what that does. I was thinking all the current at the output of the tube was due to the current passing through the tube. This seems to add to that.

 

[JohnW63]

After some staring and a bit of reading, which I must admit is rather dry reading, I have a question on the current going to the output transformer.

The voltage applied to the line that feeds the output transformer comes of the full wave rectifier as DC. But the signal coming down the line from the left, in the diagram, through R34, R35, and R36 would be AC, since it's the guitar created sound. The output transformer will only pass AC. So, should the mA be AC ? The answer is, yes it should be. I was measuring in DC. I have plenty of mA of current, now. Still just -15Vdc going to pin 2 of the power tubes, rather than -17. At least it has only one measurement out of spec. I'll follow that rabbit trail, now.

Sorry, just posting my musings in the forum.

 

[JohnW63]

I checked the ohms on all the resistors in the chain from R33, R32, R31 and the ones going to V4 and V5, R27-R30. All were where I would expect them, but, R31. It was all over the place and not stable. I just goes to ground after it, so I can't figure out why it would be wacky. Maybe I'll order another resistor. A pack of 10 from Amazon is less than a single from Mouser.com. $5.39 with Prime.

 

[DThomasC]

After some staring and a bit of reading, which I must admit is rather dry reading, I have a question on the current going to the output transformer.

The voltage applied to the line that feeds the output transformer comes of the full wave rectifier as DC. But the signal coming down the line from the left, in the diagram, through R34, R35, and R36 would be AC, since it's the guitar created sound. The output transformer will only pass AC. So, should the mA be AC ? The answer is, yes it should be. I was measuring in DC. I have plenty of mA of current, now. Still just -15Vdc going to pin 2 of the power tubes, rather than -17. At least it has only one measurement out of spec. I'll follow that rabbit trail, now.

Sorry, just posting my musings in the forum.

The instructions to set the bias to 40mA means DC current, not AC. If you can not achieve 40mA, then it's possible that the tubes are no good. Are these the same EL84b tubes that have been in the amp all along? If so, they've taken quite a beating.

R34, R35, and R36 along with C26, C27, C28, and C29 are filters for the power supply. Starting nearest the rectifier (junction of R36 and C29) there should be mostly DC voltage but a lot of AC voltage going along for the ride. As you go down the line, after going through R36, R35, and R34 to the point marked "C", there should be very smooth DC voltage with very little AC from the rectifier. Assuming C26, C27, C29 and C29 are in good condition there won't be much AC from the guitar there either.

I checked the ohms on all the resistors in the chain from R33, R32, R31 and the ones going to V4 and V5, R27-R30. All were where I would expect them, but, R31. It was all over the place and not stable. I just goes to ground after it, so I can't figure out why it would be wacky. Maybe I'll order another resistor. A pack of 10 from Amazon is less than a single from Mouser.com. $5.39 with Prime.

If you're not able to get more negative voltage than -15, then its worth a shot.

 

[Nuuska]

Hello

" The output transformer will only pass AC. So, should the mA be AC ? The answer is, yes it should be. "

Right and wrong.
Transformer passes only AC through - from primary to secondary winding.
But the bias DC-current flows only in primary winding.