Replacing PCB mounted input jack // Crate VC-2110

[Fresh_Start]

I have my first repair job (other than my own Blues Jr.).  It's a 15-year old  Crate VC-2110 with a bad input jack.  The jack is mounted directly on the PCB, and it seems like one of the pins has broken free of its solder pad on the underside of the board.  You have to push the jack over to one side and hold it there to get any sound out of the amp.  The problem's been developing slowly over time.

A couple of questions about the input jack:

1) Is there any way to replace/fix the jack with removing the entire PCB?  I can't see one but figured it's worth asking.  My experience lifting the PCB in my Blues Jr. was not great.

2) Is it better to simply replace the PCB mounted jack or use an open-style Switchcraft jack and connect to the board using short solid-core wire?

Other random questions.  BTW the owner likes the way the amp plays and sounds:

- The EL-84s are 15 years old.  The owner doesn't crank the amp much and doesn't play live at all.  If the power tube voltages are close to specs, is there any point in trying some other power tube sets?  (I've got multiple sets of EL-84s due to Blues Jr. blues)

- What are the diodes between the power tube plates & cathodes for? (D1 & D2; 1N4007)  My guess is some kind of surge supression but I don't know...

- Is there anything else that I should look out for?  15-year old electrolytics ought to be OK, right?  I'll check all of the coupling caps for DC leakage.  None of the resistors look toasted - they're all metal film except for a couple of power rail resistors.

- Any obvious tweaks that would be worth considering?  I'm not touching anything if I can avoid lifting the PCB up.  However, if I do have to remove the PCB to fix the input jack, now's the time to try alternatives.

The "Punch" switch (labelled Mid Boost on schematic) may be a bit too much of a good thing for the owner.  I don't know how to do TMB tonestack math, but the .01uf bypass cap looks potentially big and so does the 68K tone stack lift resistor.

Overall, this looks like a nice, simple circuit.  I never saw a TMB stack right before a concertina splitter before but it obviously works.

A couple more "knits" looking over the circuit:
- 100 ohm screen resistors? (same as Blues Jr.)
- no input grid resistor? (even Blues Jr. has 10K)
- why did they draw the "Punch" single-pole/double throw switch the way it's shown?

Based on the voltages shown & the cathode resistor, somebody was paying attention to the Mullard specs for 2 tubes in push-pull:
http://www.mif.pg.gda.pl/homepages/frank/sheets/129/e/EL84.pdf
DEAD on the Va = 300 specs if I'm reading it right.

BTW the first thing I noticed looking at the chassis was that there's an unused hole for a tube socket.  Inside, you can see all of the space on the PCB for the VC-2112 with reverb including labelled spots for the components.  Plus, there's even a hole hidden under the faceplate for a Reverb pot.  Anybody ever played the reverb version?

I can take pics tomorrow if that will help sort out the input jack repair options.

I'm glad there's no video of me trying to figure out how to remove the chassis!    The bottom of the speaker baffle sits in a rabbit and there are four screws on the sides holding the baffle in.  Unscrewing the chassis bolts on the back of the amp before removing the baffle is not a good idea.  (I did figure that out before removing all 4.)

Thanks in advance for any help.

Chip

[sluckey]

Is there any way to replace/fix the jack with removing the entire PCB?

Hard to say without seeing the amp. Based on what you said above, I'd just fix the bad solder joint. The jack is probably OK. You may be able to lift the board just enough to resolder the connection without actually completely removing the board.

The diodes are for flyback protection. They will prevent any negative signal excursions from appearing on the plate, mainly caused by wailing away with no speaker connected.

I wouldn't mess with anything that's not broke.

That Mid Boost switch is really wired as a SPST. That's a common way to draw a DPDT slideswitch. The horizontal lines next to the contacts are the sliders and they move left and right in that drawing. Is the actual switch a slider?

[Geezer]

Is the actual switch a slider?

Having owned several Crates (& noting the "in" & "out" on the schem), I'm guessing it's a pushbutton switch(?)

G

[sluckey]

Having owned several Crates (& noting the "in" & "out" on the schem), I'm guessing it's a pushbutton switch(?)

Ahh! Excellant point.

[Fresh_Start]

It's a flimsy looking push-button switch but appears to be working fine.

Here are 2 pics.  This is a really simple circuit - you can see the empty spaces & socket in the middle where all of the reverb (2 triodes + a transistor) would go.

Don't I need to be able to get at the solder pads on the back of the PCB to fix this problem?  If there's another way please tell me.  Otherwise, I'm going to have to take off all of the knobs & pot nuts and lift the PCB at least a little bit.

I wouldn't mess with anything that's not broke.

OK, OK... I got it!

You know I can't keep myself from at least swapping tubes, right???

Cheers & thanks,

Chip

[sluckey]

Otherwise, I'm going to have to take off all of the knobs & pot nuts and lift the PCB at least a little bit.

Yes. That looks like an easy board to pull, even if you have to totally pull it out of the chassis. I bet you won't have to do that though.

[PRR]

> prevent any negative signal excursions from appearing on the plate

Negative won't hurt.

The theory is that, by CT Transformer Action, the voltage on the _other_ plate and winding-end can't kick-up past "twice the DC voltage".

He's got 300V DC. Imagine one tube conducts BIG current and then cuts-off abruptly. (Typical gross overdrive.) If the amp is UN-loaded, the OT inductance tries to kick-up that plate to infinite voltage. With real iron-core, 10X DC voltage is possible. At 300V DC and using double-500V fishpaper, the ~~3,000V spike exceeds the ~~1,000V rated insulation.

But the voltage on the other side of the CT "should be equal". So it kicks to negative 3,000V.

With the diode, the "other side" can't kick further than from the 300V DC voltage to diode-ground, 300V. So the cut-off side can't exceed 600V. Which is safe on 1,000V insulation.

The flaw is that the coupling from one side to the other may be fair or it may be awful. And the diode is slow. It is certain that the first microSeconds of the spike will not couple across from one winding to the other, and the common silicon diodes take microSeconds to turn-on. It probably "helps some": insulation breakdown tends to be a progressive thing. Short overloads may not break down, but hold the voltage and eventually it breaks-down. The OT may be wound for 10,000 hours of easy playing, 100 hours of gross overdrive, and 1,000 hours with diodes clamping most of the peaks. Since this moves the failure from maybe-in-warranty to well out of warranty, and costs 20 cents, it is warranted.

The tone control between driver and power tube grids makes a good amp bad. It is hard to drive power tube grids. You don't put loss here.

However today if you want a "good" amp, you get something with chips. Cleaner than the best tube amp, also lighter and cheaper. Tube amps hang-on because their "bad" can be "good". Even so, you need the high sensitivity of EL84; most other power tubes, if tone stack does much (has substantial loss so you can dial-in a lot of "boost"), you won't be able to push the outputs to the full rated power you paid for because the driver will be flatlining.

[Fresh_Start]

This is off-topic, but I started the thread

If you had a cathode follower driving the tone stack or used an "active" tone stack, then the topology would be OK, right?  Or a pentode driving the tone stack?

Just daydreaming again...  Real world work is interfering with my ability to spend time with a soldering iron.  At least this repair shouldn't take long and will make a "starving student" happy when his amp works again.

Cheers,

Chip

[Fresh_Start]

Thanks to you all for your comments here.  Turned out 3 of the 4 legs on the jack were unattached!  I did have to disconnect all of the pots & jacks on the front panel to lift up the input end of the board, but didn't have to remove it completely.

Got a bit of a scare when the power light wouldn't come on after I put everything back together.  One of the quick connect leads had quickly disconnected

Voltage was high (14 volts vs. 9.? on schematic) on the power tube cathodes so I swapped in a set of spare set of EH EL-84s.  Cathode voltage dropped to 10.2 VDC.  Sounds much better to my ears than the 15-year old tubes but I don't know if the owner (a) will hear the same thing, or (b) want to pay $15 for a very slightly used pair of tubes.

Cheers,

Chip

[jjasilli]

This is a common problem with board-mounted pots & jacks, especially the input jack which tends to see a lot of use.  Use of the jack tends to flex the solder connections which get weak & fail.  Simple solution, always try keep a pedal plugged into the jack, even if it's a tuner.  Plug the guitar in & out of the pedal instead of the amp. Or, fasten a jack to the chassis and run wires to the board.