Jerry's "Gonna Rebuild me PLWO 1000" thread

[jbeckva]

Well, til then. Here's the bulk of the rework. Not too shabby. The heavy buss bars also serve to keep the sockets in line when mounting the transistors.

Oh yeah, here's another "picture worth a thousand words" (dagnabit.. they're on their way tho)

 

[jbeckva]

To ground yes.

Which ground? same as the shield.. the tabs.. the places ya said to connect the direct jumper to star ground? Or the chassis?

 

[laatsch55]

The copper bus plane underneath the RCA ins.

 

[jbeckva]

The tabs where the shields are soldered to/themselves soldered together and wired up to star ground, in other words? My RCA's are insulated from the chassis.

 

[laatsch55]

Only until they get to the Star ground.

 

[soundude]

Jer, you pulled all the green sockets back out and use the original sockets?
I used to put those sockets in when I was building regulated power supplies and I had the same issue, the transistors wouldnt go in because the solder ran into the pin holes....couldnt see any light in that tunnel...LOL So know I always go for the wafer sockets, lower profile too. I was telling Lee do yourself a favor and install those .33' caps on the back plane too. cleans up the sound from the dirty PS...rf interference too...
May the force be with you!....

 

[mlucitt]

You guys should do yourselves a favor and not bring the ground from the RCAs all the way back to the star ground. It is not necessary. If you look at Joe's board, he used R2L and R2R to keep the input grounds electrically close to the star ground without actually connecting them together. This means you can bring both input wires from the RCA jacks to the board input connectors on 1L/1R and 2L/2R with wires as short as possible or even coaxial cable.

Also, connecting the RCA shields together with that copper buss bar is destroying your channel separation. If you look at some high end amps designers often do not connect the shields. I don't know why PL did it, it would have been cheaper to run regular wires, even back then when copper was cheap.
The output shields of the preamps are normally where the signal wires are tied together so doing it again in the amp can create a ground loop if both ends are not tied back to the star grounds (which they usually are not).

I agree with not connecting the input shields to the chassis, and Joe's method lifts the inputs just enough to safely break the possibility of ground loops and hum. But I am open to experimentation and testing because I have my bench mule PL 400.

Thoughts?

 

[soundude]

You guys should do yourselves a favor and not bring the ground from the RCAs all the way back to the star ground. It is not necessary
I agree, your input grounds should be brought back to the input stage and not combined with the PS common, the input circuit could possiblly be resistive to the PS ground and connecting it to the PS will defeat the purpose of seperation / isolation and could reproduce a 60hz injection.

some high end amps designers often do not connect the shields

This is a sensitve subject because if the rca input shields are isolated from some sort of ground or negitive connection the amp would produce a phantom signal close to that of the surrond sound like a theatre system. If the input signal section of the gain stage is completely isolated from each other and the rca inputs are isolated from the chassis there needs to be some sort of returning path. Most of what Ive seen some high end designers use the chassis as the common since they isolate the utility ground to a isolated stake point and couple it to the system ground through some sort of resistor or coupler cap or even a r/c network. If the shields are not connected then the rca are usually common to chassis.

I actually perform this on my phono amps and some preamps, if the case is common the system ground it becomes a excellent shield for rf especially in high gain situations...

 

[soundude]

I don't remmber if I posted this in this forum but there is a excellent book called the radiotrons designers handbook 4th edition, unfortunatly it was published in 1952 so alot of it is valve related but it has a excellent powersupply and grounding section that would pertain to al sorts of amplification. For you toobies this is a must have....
Here is the site, its PDF by chapter so there is alot of downloading but worth its weight....this book in original form is going for crazy money, it is considered the audio bible....
http://headfonz.rutgers.edu/RDH4/

 

[jbeckva]

Ok, so here's where it gets confuzzlin' a bit. At least on my WO driver board, R2L is 2.7 ohms while R2R is 56 ohms. So isn't that an imbalance in the ground lift between channels? Is that difference between L and R part of "current" design or is it now better practice to have both of them at 2.7?

Should I really separate the two ground/shield connections at the RCA's and take away the jumper to star ground? In previous experiments I had to have that jumper, or the hum was unbearable. BUT.. I also didn't have the two 47K's in parallel with the inputs either, so perhaps that is what is needed.

I got Lee's "picture", but I'd like to see others...

 

[soundude]

Thats a question for Joe, don't know the circuit layout. But by process of elimination u can isolate the problem. If I understand you correctly ur going from ground or chassis? Ground to rca input with pots fully counter clockwise will give you the resistance of the pots at min. position, I've noticed that the original PL pots don't directly short the inputs. you can confirm this by shorting the pot connections, shield to both of the (think) red wires soldered to pots, then take you readings again and they should be 0. I too have gotten 2.7 ohm and different on other channel and narrowed to the pot wipers not directly connecting to the shield termination of the pot..

 

[laatsch55]

Jer, drop the 56 ohmere and go with a 2.7 there.

 

[soundude]

Jer, the 47k's are only needed if you connect directly from input jacks to Joes board, meaning no volume control in circuit. you still have your direct / normal switch rite? No need, Lee added another set of inputs under the originals set, which is his direct and the others are controlled by volume or gain controls on front of amp. controls act as the resistor and keep the original inputs in a closed loop with resistance. His added inputs were direct with no control and open looped so it acts as a antenna and picks up any kinda noise around the circuit, the 47k closes the loop so the noise is refrenced to ground and the system becomes dead quiet....

 

[laatsch55]

Gonna be gone for a bit today. Taking one of the Grandaughters to Cabelas in Rapid City to complete her fly fishing ensemble for her birthday. 10 years old and absolutely LOVES to fish. otta encourage that all we can.

 

[laatsch55]

If you'll notioce, the original PL 14 and 20 boards were grounded from the speaker outs which were grounded by the chassis via the 3 place terminal strips on the back wall, THAT I never understood.

 

[laatsch55]

Jer, that 2.7/56 ohm difference is nothing more than Bob playing around onew day and stopping the production line saying the 56 sounds better there, so start putting them in there. True story.

 

[jbeckva]

Hmmm... Ok, so first I am going to put a 2.7 on the right side.

Now, what I have right now is from the RCA's (I only have one pair... no switch), wired to the pot. Originally yeah, the pots were backwards but I have that corrected with 100K across the inputs "constant" because the input high and shields are connected to the two pins besides the wiper, and the wiper goes to the WO board input. Shields/grounds ... one from the RCA side .. the other to the WO board input.. are connected together and wired to one of the two constant/non-wiper pins on the pot.

Now up at the RCA's, the two tabs that connect the outside rings of the RCA's to the shields on the lines going TO the pot are soldered together, and a wire run from that connection over to star ground.

My original plan was to break and separate the two L and R input lines (those from the RCA's to the pots), and connect at the WO board "normally" i.e. the high side from the pot's wiper to the high side input.. the shield from the pot to the shield on the WO board. That didn't do too good... got a lot of noise still sticking around.

BUT also tho... in reading a few recent posts, I recall seeing one that said the pots themselves via the front panel really need to be grounded, eh?? Since I don't have the nuts for these ALPS right now, they're kind of hanging free. Is that the real reason for my noise issue?? Hmmmm...

(can only speculate right now.. dagnabin tranny pin... still have to wait.. )

 

[soundude]

Lee, Is this a resistor on the board?
I know the pots don't short perfectly...
Have fun fishing, wish my son liked fishing...

 

[laatsch55]

Not going fishing yet John, just getting the Lass outfitted. Yes the 56 ohmer is on the board, and causes unequal gain issues. Run a terst lead Jer and ground pots to faceplate, if there is no connection there now.

 

[jbeckva]

I need mah nutz.... LOL (lee knows...)

or if you can tell me what exactly to go look for at Lowes or Tractor Supply, I can get 'em ...