Paramour 1 low output problem

[ssssly]

I have a pair of Paramour 2s and a Foreplay 2 that I built about 7 years ago and have worked wonders for me ever since. I recently returned home from an 8 month deployment, and needing some good relax time, fired up my system. There was a huge buzz out of the right speaker and I immediately shut everything down.

Upon inspecting the Foreplay the problem became apparent. The Capacitor connected to from the RCA jack to B8 had managed to work its way loose (actually it pulled the pin out of the tube base: guessing during one of the earthquakes while I was gone).

So I hooked up this Chinese tube preamp I picked up off of ebay some years back and to my disappointment the paramour that was hooked up to the affected Foreplay channel now has almost no output. The volume is very low, almost inaudible compared to the other amp, and the sound is very distorted. I checked the voltages and they seem to all be fine, within 10-15%.

Any thoughts?

[Paul Birkeland]

Try swapping the tube sets between the two amplifiers.  I would bet that you have a 12AT7 or maybe an earthquake shaken 2A3 that has gone bad. 

[JC]

Not to be too pedantic, but you have swapped signal cables from the preamp, right?  I just would hate to see you chase your tail on the Paramour, only to find out that the other preamp or the interconnects are your problem.

[ssssly]

Tried that. Swapped tubes, speakers, speaker wires, RCAs. Is definitely something with the amp.

[ssssly]

Stripped the amp dow and rebuilt it replacing everything, with parts on hand, but the Parafeed Coupling cap and the 15k ohm wire wound for the c4s (didn't have any caps rated 600v laying around or the wirewound). The wirewound measures good on my DMM. Everything else I put back to stock values with new caps and resistors per the original manual.

The C4S LEDs are not lighting up (they were before the rebuild) and the measurements are as follows.

Resistances:

Input Jack:  269.5
1)   220.2 ohm
2)   207.5 ohm
4, 12, 14) 0 ohm
17)  270.9 kohm
25)  182.1 ohm
A1)  1.195 kohm
A3)  270.4 kohm
A4)  1.195 kohm
B7)  270.1 kohm
B8)  497    ohm

2 and 25 are a bit low but don't look outrageous to me.

Voltages:

1)   311.7 vac
2)   312.0 vac
5)   380 vdc
15)  366 vdc
A1)  60.5 vdc
A2)  353 vdc
A3)  140 vac
A4)  60 vdc
B1,2, 3) .04 vdc
B6)  254.5 vdc
B8)  2.6 vdc
Left Most Terminal as in C4S manual) 252.2 vdc

Other than the high B6 which is normal on the V1 C4S the voltages look ok to me other than the B1-3 voltages (not sure what they are supposed to be for a C4Sed mour.

No idea why the LEDs aren't lit?

Might replace the Parafeed Cap with a 400V just to see if that is causing some issue.

Oh, and for some reason the on/off switch is now reversed. What was off is now on and visa versa.

Thoughts?

[Paul Joppa]

Do not replace the parafeed cap with anything rated less than 600 volts! It cannot cause any of the symptoms you have described, but it will be damaged if a lower voltage rating is used, and it might take out the output transformer when it fails.

The RCA input resistance should be 270K; you measured 269.5 ohms - I assume that is a typo? You got the correct value at B7 which measures the same resistor.

A3 should measure near zero DC and small AC - perhaps you measured 140mVAC not 140vAC?

I can't think of any way to reverse the power switch unless it were rotated 180 degrees or a different switch was used.

Now you measured 2.6 volts at B8, indicating 5.2mA current. That suggests the LEDs have too large a voltage drop - this has been a problem for several people in the past. The easy fix is to replace one of the two 499 ohm resistors on the C4S board with something larger; usually 750 ohms has worked well but 820 or 680 are close enough. It is also possible that the 12AT7 is worn and needs more voltage than usual to draw that much current. The combination of these two problems has in the past reduced the "compliance" - the voltage difference between B6 and the supply to the C4S board (you measured a bit over 250v for both). When this compliance falls below the transistor saturation of a few volts, the LED bias current can get drained off through the transistors causing them to get dim. Usually one is dimmer than the other, and in bright light it can be hard to see if there is any glow at all.

If there is really no glow at all, then I would have to guess that the B8 measurement is in error, perhaps being mV instead of V. In that case I would suspect the B6 socket pin connection to the tube pin, or the very difficult magnet wire that connects B6 to the C4S board may have a bad solder joint.

Those are my best guesses right now. The other voltages and currents appear to be correct.

[ssssly]

Found the C4S problem. Had a couple of wires touching. Original problem still remains. Output is less than half the other amp, is easily distorted and there is quite a bit of hum (almost equal to the output volume).

New measurements are:

Resistance
Input)  269.1 kohm
1)   219 ohm
2)   206.5 ohm
4, 12, 14) 0 ohm
17)  270.9 kohm
25)  181.6 ohm
A1)  1.195 kohm
A3)  270.4 kohm
A4)  1.194 kohm
B7)  270.0 kohm
B8)  497.0 ohm

All well within specs

Voltages:

1)   269.1 vac
2)   269.4 vac
5)   330 vdc
15)  316.9 vdc
A1)  51.2 vdc
A2)  306.5 vdc
A3)  125 vdc (have an auto ranging DMM triple checked and it is reading vac not mvac)
A4)  50.9 vdc
B6)  246 vdc
B8)  2.08
Left most term) 226.4 vdc

Wasn't planning on leaving the 400V cap in as a fix just to see if by some bizarre chance it made a difference.

Don't have a 820 or 680 ohm resistor on hand. Are the variations seen there the possible cause of the issues I'm having? They don't seem that far out of the norm for others I have seen posting without problems. 

[Paul Joppa]

These voltage measurements look to me like they indicate a few problems still. I'll describe what I see from the problem measurements. This post is long because I wanted to describe my reasoning.

1)   269.1 vac
2)   269.4 vac

These are the power transformer secondary, and should be 330v each. If this is accurate and the transformer is working correctly, it indicates your power line voltage is a bit less than 100 volts, where it should be 120 volts. Are your lights very dim?  :^)  Seriously, I would make sure about this first, it seems on the face of it unlikely but could explain all the problems. Check the power line voltage, and then the filament voltage (AC) across each tube - between A1-A4, and between B9-B4/5. And if you haven't noticed a 20% low power line voltage already with other appliances, then make sure your meter is working properly!

5)   330 vdc
15)  316.9 vdc
A1)  51.2 vdc
A2)  306.5 vdc

All the above voltages are 20% low, consistent with a 20% low line voltage.

A3)  125 vdc (have an auto ranging DMM triple checked and it is reading vac not mvac)

The A1 and A4 voltages indicate the 2A3 is conducting properly, so this voltage cannot be 125vDC - the DC component must be quite small. If it is really reading 125vAC then it may be oscillating badly. (In that case, it's also possible that the meter actually measured peak voltage on teh DC scale, and would read a DC voltage when there is no such voltage on an average basis. This is unlikely but not unheard of.) Make sure the corrected grid stopper wiring has been installed - there should be a 270K resistor from T17 to ground at T14 (NOT A3 to ground as shown in the original manual) and a 499 ohm resistor from T17 to A3. Nothing else should be connected to A3.

It is conceivable that if this is oscillating badly, it could bring the whole power supply voltage down, and the 2A3 would be getting quite hot.

A4)  50.9 vdc
B6)  246 vdc
B8)  2.08
Left most term) 226.4 vdc

The "left most terminal" cannot have a lower voltage than B6, so the voltage must have drifted while you were measuring. It may indicate your power line voltage is fluctuating as well as being low, or it could indicate destructive oscillations that are heating or otherwise causing components to drift.

The fact that this voltage is about the same as the driver plate (B6) indicates the C4S is not able to force its target current through the tube so there is no compliance for positive peaks at the driver output. This would cause severe distortion and low output, it would also couple power supply ripple to the 2A3 grid and cause a good deal of hum.

I mentioned a few possibilities in my previous post, but here are two new ones: A) the left most terminal should be at 295v, giving it a much greater opportunity to exceed B6 by the minimum of 60-70 volts, and B) if the heater voltage is also 20% low the 12AT7 cathode may not be hot enough to provide the necessary current.

Sorry for the long post and many options. I hope it gives you some things to check that will lead to a solution.

[ssssly]

Thank you much Paul. I'm gonna take a day to digest that and see if I can borrow a multimeter from someone so I can set the range manually.

I do have a bit of fluctuation in my power here (Okinawan power grid) but I have the line that goes to the listening room on a step up transformer (100-120) and my power conditioner reads fairly steady at 120. Actually oscillates up to about 123 but very rarely drops below 120; even under load from my system at rather high volumes. As long as I am not drying clothes the power is passable. I will make sure I double check when I take the next batch of measurements. 

I'll recheck the wiring (not sure which manual I have) and see what I get.

Is there any way to get the Paramour and Foreplay 2 manuals on PDF? Took me hours to figure out where I put them and I have gotten used to working out of my E-Book.

Thanks again for everything.

[ssssly]

Line voltage is definitely 120.

I apparently have one of the original manuals as I had the 270k resistor to A3 not ground. I changed that as per your instructions and have the 270k going from T17 to T14 and a 499ohm resistor from T17 to A3.

With that change the voltages are as follows.

A1-A4)  2.4 vac
B9-B4/5) 5.8 vac

1) 314 vac
2) 313 vac
5) 394 vdc
15) 377vdc
A1) 62.6 vdc
A2) 363 vdc
A3) 37 mvdc
A4) 62.8 vdc
B6) 223 vdc
B8) 2.35 vdc
left most term) 273 vdc

It is playing music again. The volume is still lower than it was (as well as the other amp) and a big enough difference to make listening a bit of a chore. Female vocals particularly. Seems there is a sweet spot in the output right in the female vocal range where the output is equal creating a panning as the image shifts back and forth from the center to the left speaker. But at least I don't think I blew up anything expensive. Might just need to burn in a bit, new caps and all.

Is the difference between B6 and the LM Term enough or should I get some resistors for the C4S board and try to broaden the gap a bit still?

Thanks

[Paul Joppa]

Excellent! Looks like it was indeed oscillation, which is the reason for grid stoppers. It's always hard to predict, being sensitive to the exact wire locations. I believe this is the first time I've actually heard of this wiring error causing such oscillations in a Paramour, though it was always a theoretical problem and I have had it happen in a different 2A3 amp so I knew it was possible. Make this change on the other amp if you haven't already!

Yes, the B6 voltage is too high. You need a small margin (5-10v minimum) beyond the A1/A2 voltage (63v in your measurement) for a total of about 70v; you have 273-223 = 50v which will limit the peak output by about 3dB (half power). Just replace one of the 499 ohm resistors on the C4S board with 750 ohms approximately. You'll want to do this on both amps to maintain symmetry. This change should get the B6 voltage below 200v, and the LM terminal voltage should rise a bit as well.

In our original build as described in the Paramour C4S Upgrade manual, we got 155 volts at B6. This does vary a lot between tubes, but it also rises as the tube ages. It is possible that the reduced gain is a result of the 12AT7 being worn - I think you said you've been using them for seven years? The low measured heater voltage could also cause reduced emission, especially with an older tube. I can't see a reason for that voltage to be low when the primary and high-voltage power are spot on, though some meters are less accurate at low voltages. The 2A3 filament is pretty dim and often hard to see in a well-lit room, but you should see an orange glow from the 12AT7 heater pretty clearly. You might swap tubes between channels since the other monoblock seems to be working fine.

The bias point of the 2A3 is spot on so I don't think it is running down at this point.

One other slim possibility, but worth checking. Your description of a wandering image could be due to a phase reversal between channels, i.e. in the speaker cables. (It's not unknown for speakers themselves to be phase reversed, though pretty darn rare.) It's just a thought; I can't think of a reason for noticeably different frequency response between two properly operating Paramour amps.

[ssssly]

Finally got some parts in. The following are the reading with a 680ohm resistor in place of one of the 499ohm resistors on the C4S.

1) 320 vac
2) 320 vac
5) 392 vdc
15) 377vdc
A1) 62.2 vdc
A2) 362 vdc
A3) 41 mvdc
A4) 62.1 vdc
B6) 204.9vdc
B8) 2.05 vdc
left most term) 281.3 vdc

With a 750ohm resistor it drops the B6 voltage to 94 and the left most term rises considerably. Don't remember what it was and for some reason I didn't write it down. Oh, and the 2A3 got rather hot considerably faster than it normally does. Don't have anything than my subjective timing to go by but I almost burnt myself taking it out after doing the voltage readings and have never had that issue before with doing voltage checks.

Judging from previous posts I'm thinking while not optimal 204v is better than 94, giving me a 76v difference, so I will leave the 680ohm in. Please correct me if I am wrong.

And while I have had the amps for 7 years, the stock 12AT7s are not still in them. That being said, I don't have another good set on hand and have had this pair in for a couple hundred hours; so that is not to say that they may not be getting a bit worn as well.

Thanks again.

[Paul Joppa]

Yes, the 204v is fine. The 2A3 measured voltages are still spot on, so you should be good to go.

[ssssly]

Did a little more fiddling around and came up with some interesting results.

When I went to change the resistor on the other amp I checked the voltage and the B6 was right on (156).

So determined to get them as close to one another as possible I started playing with the resistors on the C4S board on the other amp.

With a 680ohm and 820ohm I had for another project I got the B6 down to 159 left most term 302.

What would cause the need for that much more resistance in the C4S on that amp?

Is there something else I should be looking at?

[Paul Joppa]

Two things affect the plate voltage:

1) the C4S current, which depends in turn on the resistance of the two parallel resistors and on the voltage of the LED. The LED voltage is not a specified parameter; it's usually around 1.57 volts for the HLMP-6000 which we use now, but does vary especially with different "red" LEDs depending on the specific chemistry of the semiconductor doping.

2) Each triode in the tube is a little different, and the C4S load does not rpovide teh DC negative feedback to stabilize the operating point (the output tube does use a resistor cathode bias to stabilize its operating point). A 30% variation in plate voltage from this cause is not uncommon. That's way better than transistors which might have a gain of 30 to 200, which is also sensitive to temperature and current, but it's not like a precision resistor either.