Oliver PA-100 Amp.......

[Jack_Hester]

Now that the G-300 is off the bench, and only drawing clean up to do, I put the PA-100X up there.  It had one bad 7027A, so I borrowed both JJ 7027A's from the 300.  The 6AN8A and the two 6EU7's were good, as well as the 5AR4.  So, all were installed and my bias test adapter installed on the power tubes. 

Did not take a complete set of voltage readings, to begin with.  But, I noticed that the B+1 was lower that of the 300  The secondary AC of this PT was right at 418v, and the B+1 489vDC.  The 300 had a secondary AC of 413v and a B+1 of 505vDC.  The difference may be due to this rectifier tube. 

Anyway, my bias voltage leaving the diode is -35vDC and my bias current is right at 100mA.  I won't get back to this amp until the middle of the week, but I think I'll swap out the 5AR4 with another one, just to see if my B+ increases.  And, do a bit more troubleshooting on the bias circuit. 

Other than these issues, the amp is quiet.  I didn't bother to connect the guitar, as of yet.  I'm wanting to lift one side of the 47K resistor in the bias circuit, and measure it's value.  Hopefully, I can get a measurement on the 56K, but it's too buried to lift without lifting other components. 

I'll post more when I'm able to take more readings.  Thanks for any suggestions and comments.  Have a good one. 

Jack

[sluckey]

That schematic has a big error. There must be a center tap on the HV winding. Otherwise, you can't make any B+ or bias voltages.

[Jack_Hester]

That schematic has a big error. There must be a center tap on the HV winding. Otherwise, you can't make any B+ or bias voltages.

I agree, and I confirmed the CT on mine, while I was making comparisons with the G-300. 

Jack

[Jack_Hester]

Haven't been able to do the troubleshooting that I had hope to have done by now.  Maybe tomorrow (Sat) after I get some seriously deep grass cut.  It's calling for an all day rain, though.  I don't need that, just yet.  But if it comes, then I'll get in some troubleshooting on the bias supply. 

In the meantime, if someone would, please do me a favor and do the math on this circuit.  The PT secondary is still right at 418vAC, and the bias voltage measured -37vDC.  It occurred to me to look at my schematic of my Ampeg B-12-XY, as I made a voltage chart for that one.  I had a PT secondary of 417vAC, an identical bias supply circuit (34K instead of the 47K), and -52vDC out. 

If the rain sets in, I'll see if I can de-solder the components and take some resistance readings.  The diode is shaped like a big round pill, but if need be, I have plenty 1N4007's to pick from as a replacement.  It's labeled as SD-1, and is extremely easy to get to.  Along with the 100K.  The 47K is kinda buried under the big fat 100uf-100v cap, and the 56K is completely buried under a big ceramic block 20K resistor. 

Anyway, I'll see what the morning has to offer.  More to come.  Thanks for any help.  Have a good one.

Jack

 

[Jack_Hester]

Here's the reply that Dave H. gave on the Music-Electronics forum:

I’d normally just simulate it but I’ll have a go.
 
 R1 and R2 form a voltage divider with an output resistance of R1//R2 so with 418V input the divider output voltage is 150V with an output resistance of 35.9k. The 150V is then divided down by a second voltage divider formed by 35.9k and the 47k (R3) so the voltage is 150 * 47/(39.5+47) = 85V but the cap only sees half of this because of the diode ie. 42.5V. Assuming the 418V is RMS we need to multiply 42.5 by 1.11 because the cap will charge up to the average value not the RMS so we now have 47V.
 
 1.11 is the form factor of a sine wave which is RMS/Average value. For a sine wave RMS = peak/sqr(2) and average = 2/pi if I remember correctly so that’s 0.707/0.637 = 1.11

I would love to see comments from others here, and see if anyone has another take on this.  I will proceed from here and post my troubleshooting, as it evolves.  Have a good one. 

Jack

[Jack_Hester]

Spent some time playing with the math.  Then, I rolled it into a very small program, to play with it some more.  Requires no installation.  Just extract and run.

Jack

[HotBluePlates]

Did not take a complete set of voltage readings, to begin with.  But, I noticed that the B+1 was lower that of the 300  The secondary AC of this PT was right at 418v, and the B+1 489vDC.  The 300 had a secondary AC of 413v and a B+1 of 505vDC.

I don't know what "B+1" is. If the two amps have different power transformers, rectifiers, filter cap values, filter cap newness, or circuit current draw then the voltages won't necessarily match.

And when you mention "low B+" the first thing I saw in the schematic is the 1kΩ resistor that all the amp's B+ current passes through. But given your questions about bias, I don't think you want to add B+ voltage at this moment.

Anyway, my bias voltage leaving the diode is -35vDC and my bias current is right at 100mA.

In the meantime, if someone would, please do me a favor and do the math on this circuit.  The PT secondary is still right at 418vAC, and the bias voltage measured -37vDC.  ... my schematic of my Ampeg B-12-XY ... had a PT secondary of 417vAC, an identical bias supply circuit (34K instead of the 47K), and -52vDC out. 

Here's the reply that Dave H. gave on the Music-Electronics forum:

I didn't open your math program, and you don't want to calculate. With due respect to Dave H., the math for a rectifier circuit very painful and is not as simple as what he presented. Ohm's Law stops working with the non-linear process of rectification, and every part in the circuit has an impact on the answer. The complexity is why you never see "gurus" present a simple formula for computing B+ voltage with a given a.c. input and rectifier tube, because there's so many factors involved and steps to do (you can find them in RDH4 if you really want).

So instead, let's look at your amp's bias circuit, and see what parts will change its output voltage. It will be faster for you to tack-solder parts and see what happens than to calculate anyway.

If the 0.1uF coupling caps going to the output tube grids are leaky, they will drag the bias voltage closer to 0v. So pull the output tube from the amp, disconnect the wire from the bias supply to the 120kΩ bias feed resistors, and measure the negative voltage on that wire. If it has gone up, you know the coupling caps need replacing.

If your 100uF cap is old and leaky (or has reduced capacitance-value), it will drag the bias voltage down. If you're doing frequent repairs, you have something similar on-hand (or should)... Swap the cap and see if the bias voltage goes up. If it does, the old cap needed replacing anyway.

Dave H. pointed out there is a voltage divider from the high voltage to the diode to reduce the raw a.c. input. There is also a 47kΩ after the diode to again reduce rectified voltage. If none of the above steps raised the bias voltage (because the caps were good), then you have a choice: you can raise bias voltage by fiddling any/all of the resistor values, but which one?

If you don't see signs of damage, and didn't notice low resistance values for any of these resistors while the power was off, then you can either raise the value of the 47kΩ, raise the value of the 56kΩ or lower the value of the 100kΩ resistor. Any/all of these changes will increase either the raw a.c. input or the rectified voltage. The fastest way for you is to lower the value of the 100kΩ. Get a 1MΩ pot, and use 2 of the terminals to create a rheostat, which you will tack in parallel with the 100kΩ. With the amp off, set the pot for maximum resistance between the terminals you used. With power applied, carefully turn the pot and monitor bias voltage, tube current or both, until you reach your desired idle current/bias voltage. Turn off, remove the pot from the circuit (without disturbing the wiper position), measure resistance between the terminals you used. Find the nearest standard-value to that resistance and solder that resistor across the 100kΩ. Bias done.

If you have to choose between higher & lower standard values, the lower value will result in a higher bias voltage and lower tube current (maybe the safer choice), and the higher value will result in less bias voltage and higher tube current.

You can calculate bias circuit values if you like, but I don't think you'll find a faster or cheaper approach than the one above.

[PRR]

> the 56K is completely buried under a big

If you doubt the parts, but they are "impossible" to get at.... something this simple, you could hay-wire a duplicate with known-good parts and see what comes out.

[Jack_Hester]

HBP -

I have the 300 buttoned up, and it's going to Church this morning, for my buddy Don to take home.  But, I did notice that the PT's and the OT' are the same for both.  Not having his other models to compare, it makes me think that he standardized for a while, on his components.  At least on these two.  That's why I was comparing the secondaries of each. 

The math was a fun exercise for me.  I'm not good at it, and still love to try and figure it out.  And, I'm with you on it being painful in something as simple looking as this.  But now that it's done, I won't take the values generated by it at face value.  Though they do make for an interesting means to experiment with changing resistance values and watching how they affect the the calculated bias voltage.  That's what I wanted to visualize, without having to give the calculator a workout (and my brain) each time. 

The 100uf is very easy to get to.  At least, one end of it.  I think that I can get to the other end to de-solder, with a little difficulty.  I have a replacement. 

Also, the 100k is easy.  And, the pot suggestion is equally as easy.  I'll strap a 1M across the 100K, and go from there. 

I would rather find the problem, than leave things as they are, and substitute.  You've given me some good starting points.  But, once that's done, and because Mr. Oliver left plenty of holes (covered with shiney chrome plugs), I think I'll still add a pot (say a 10K or 25K, as I have those) in series with the 47K.  Just to make the bias 'tweakable'. 

PRR -

I came close several times to putting this circuit together on the breadboard.  Using my existing PT, I have enough vAC to duplicate the input like I have on this amp (100X).  Nothing like the real thing. 

I'll have to do most of this next week, after work (and mowing the pasture that my yard has turned into).  But, I'm well prepared.  Thanks, and keep watching. 

Jack

[Jack_Hester]

If your 100uF cap is old and leaky (or has reduced capacitance-value), it will drag the bias voltage down. If you're doing frequent repairs, you have something similar on-hand (or should)... Swap the cap and see if the bias voltage goes up. If it does, the old cap needed replacing anyway.

You nailed it.  I removed the 100uf-100v cap, and lifted the diode end of the 47K resistor.  It measures right at 47K.  I was now able to get to the 56K to take a reading.  It too was right at its correct value.  So, I soldered the new 100uf-150v cap that I had, along with the 47K.  Reconnected the power and switched the amp on, monitoring the bias current and voltage.  Right off, it stabilized at -57vDC, and the bias current at the tubes was about 20mA. 

So, I unplugged one of the holes that Mr. Oliver provided for me, and installed a 25K-2W pot.  I removed the 47K and connected one outside leg of the pot to the signal ground, right where the 47K had been connected.  I installed a 24K resistor in series with the other outside leg of the pot and connected it to another wire going to the junction of the 100uf cap and the diode.  I soldered a jumper from the wiper to the outside leg of the pot that the 24K was connected.  Note: both wires had been twisted and routed up tight against the chassis.

I rolled the pot around so that the full 25K and the 24K were the total resistance.  Fired the amp up, and monitored the bias current and voltage.  I then rolled the pot until I had 45mA bias current, and my bias voltage at the diode was -51vDC. 

I'm pleased.  Tomorrow, I will warm it up for a short while and take a full set of voltage readings.  I will hopefully be able to give it a try with the guitar, but I won't rush it.  Only a short amount of time each evening to dedicate to it. 

I'll have more for the next couple of days.  Have a good one.

Jack

[HotBluePlates]

Glad to hear it worked out well for you!

[Jack_Hester]

I managed to get some voltage readings and make a chart.  Last night, the bias current was split some.  Tonight, it had more runtime, while I recorded the voltages.  Now, the bias currents are no more than .5mA apart.  These are JJ 7027A's that I picked up on eBay as a matched quad.  The seller said that they came out of a new amp, and he had his own preference to use, which is why he took them out. 

I put the original Mullard 5AR4 back in.  That cap in the bias supply must have really been dragging my PT Secondary down. 

Grabbed the guitar and gave it a tryout.  Channel 1 is some kinda loud at half volume, and very clean.  This is with the guitar volume at less than half.  I'm using the same 12" Eminence Wizard that I used on the other Oliver.  Channel 2 is just a little tamer, but not by much.  And, equally as clean as the other. 

I noticed on the schematic that there is a pot on the PT Primary winding, and one on the 6.3v Filament winding.  I didn't see but one, but I'll have to flip the chassis back over to determine which one that pot is.  It doesn't seem to make any difference in the hum, that I can tell. 

I will be making a new schematic of this amp, with layout, at some point.  I'm already seeing differences in the power supply.  Component values, for one.  The OT center tap comes straight off of the rectifier, and the screen grid resistors get their supply after the voltage (B+2) is dropped.  The schematic shows the CT and the screen grid resistors coming from the same point.  I'm sure I'll find a few other differences. 

All in all, this is turning out to be another really nice amp.  I'll snap a few pictures before the end of the week.  Have a good one. 

Jack

[sluckey]

I noticed on the schematic that there is a pot on the PT Primary winding, and one on the 6.3v Filament winding.  I didn't see but one, but I'll have to flip the chassis back over to determine which one that pot is.  It doesn't seem to make any difference in the hum, that I can tell.

The pot on the primary does the same as the 'ground' switch on a Fender, only it's variable rather than switching the cap directly from one side of the line to the other. You don't really need that pot or cap when you correctly install a 3 prong power cord.

If you have a sensitive meter you can set it to measure AC volts and connect probes across the speaker. Turn all volume controls to zero. Now adjust the 'hum' pot for minimum AC volts on the meter. Should be about mid point of the pot rotation.

[DummyLoad]

just out of curiosity, what are the voltage readings on top of the 30uF and 40uF caps?



--pete

[PRR]

> there is a pot on the PT Primary winding

They must have had some really bad power in Westbury!!

Actually, that type connection might be of slight use at *my* house. Because the wire to the street is very long, and the "ground" here is high rock, electrostatic neutral is not quite the green or the white wire. However taking wall outlet Green has always been close-enough for all practical purposes. And far safer than floating things on a pot (which does nothing for safety).

Yeah, 3-pin cord, green SOLID to chassis.

I'd clear-out the pot and cap, but if you really think it affects collector value, you might leave them in. With a solid 3-pin, they do nothing.

[Jack_Hester]

just out of curiosity, what are the voltage readings on top of the 30uF and 40uF caps?

Pete -
 
I have not confirmed that those cap values are being used.  But, on my voltage chart, the voltages are my B+1, 2, 3, and 4.  B+1 going straight to the OT center tap.  B+2 going to each power tube screen grid resistors.  B+3 going to the 6AN8A tube.  And, B+4 going to both 6EU7's. 
 
Jack

[Jack_Hester]

Sluckey and PRR -
 
This evening, I'll confirm the function of this 'stray' pot, as to being tied across the PT Primary, or as a Hum Balance pot (like his Portaflex design).  I have not seen a second pot, but depending on the use of this one, I'm not above adding one or the other to one of the plugged holes.  I would prefer to have a Hum Balance across one of the power tubes, as that's where the first filament tie point is located.  We'll see. 
 
Fortunately none of the mods that I have, or will make, will modify the originality to the point that they can't be easily undone.  I'm beginning to see that it's possible that each amp produced has a slight variation from what I'm also thinking are generic schematics.  I know that my Ampeg B-12-XY had some major variations from the schematic that I received from the new owners.  If it had an original, it was lost.  A spring tank hole occupies the spot on the platform, where one would have normally been pasted.  That's why I made my own, and put a copy in the cabinet. 
 
Also, I would like to add a thermistor to the input line, but I need to check the current draw.  This amp has a 4 amp fuse, and the fuse plug is labeled for 4 amps.  My thermistors are only rated for 3 amps.  The G-300, with it's additional tubes, only had a 3 amp fuse, labeled the same.  I did not check the current draw on that one, and I don't have it here at the moment. 
 
Jack