Discussion in 'Fisher' started by Sam08861, Jun 11, 2018.
Good suggestion rider. If I go forward with "take 3", I'll add that picture too.
Here is the underside of the Unity Gain board I built with annotations. If you view this image along with the images in post 12 and the thread referred to in that post, maybe someone can tell me what I've done wrong to cause distortion when I installed the board into my 500-C.
Note that there are a few places marked "GROUND". They are all connected via the grey wire that runs vertically on the underside of the board.
I apologize for the scribbling mess. My editing program doesn't allow me to go back in time to erase stuff (I can erase only the LAST edit). I had a hard time figuring out how to represent the fact that some components are connected in parallel such that the non-ground leads are connected to each other. They are not connected to a single wire that runs across the board as it does on the schematic.
I will post a picture of how I wired the board the next time I open the receiver. It's fairly easy to describe. I disconnected the wires from the 2 center posts on the volume control. Then I took the LEFT IN and RIGHT IN wires from the board and connected them to the center tabs. I connected the wires I disconnected from the center tabs to the LEFT OUT and RIGHT OUT leads coming from the board. The diode lead from the board was connected to the same place at which I connected the IBAM board to get current - pin 4 of V14 which is where the SDS Labs board filament wire is connected i.e. the B- (I think). The ground lead from the board was connected to the same ground I used for the IBAM board which I think is just a lug on one of the terminal boards on the chassis. I tested with a DMM to ensure there was continuity between bare metal on the chassis and the ground on the board.
UPDATE Jun 16... I've added the annotated TOP of the board so one doesn't have to go to an earlier post, back and forth, to match components.
When the weather gets worse or we get a rainy weekend and I'm not compelled to be on my boat, I'll start my build. Hopefully you'll get it figured out by then
If not, I'll post as I go and we'll see if we can't get this to work.
Thanks, Sam. In the meantime, I'm going to try a variety of tasks to see if I can figure it out. Thorne
Do you have the data sheet with the LF353N pinout? In my limited experience the pins have been odds on one side and evens on the other, but that said, it's been very limited experience with ICs.
Edit, never mind, this looks correct as well.
Thanks Thornev. Only thing I can think of is that you've jumped (connected) pins 1 and 2 and also 6 and 7 as I would have too from reading the schematic on the prior page. Wondering if this doesn't negate the internal circuit of the IC and perhaps wasn't meant to be interpreted that way. Maybe that's something to try. (removing the connections between those pins)
Sam - I've attached the schematic for this mod. See if you still think those pins should be separated after you look it over. Thorne
Yep, that's the one I was looking at. I'm not sure, but what I'm thinking is that logically, by connecting 1 and 2, the internal 1 to 2 connection is bypassed. Same goes for 6,7.
Therefore, I was thinking that by removing the connection on the pins, one re-introduces the internal connection into the circuit and perhaps the schematic is showing this internal connection, but being a novice am not sure.
My thought was it might be easy enough to try and see if it works.
PS, hopefully that makes sense to you and if not I can try to explain better.
My lay theory is that the connections on the schematic between 1/2 and 6/7 are the internal connections within the chip and not at the pins (external) and so don't need to be connected at the legs.
If I disconnect 2 from 1 and 7 from 6, then 2 and 6 aren't getting connected to by anything. Does that seem to make sense?
Just read your last reply. I see what you're saying. Hmmm. Yeah, it almost looks like an infinite loop having 2 connected to 1 and 6 to 7. 2 and 6 are INputs according to the datasheet. That would mean that because 1 and 7 are OUTputs, 1 goes into 2 which outputs through 1 which goes into 2 which outputs through 1 which goes into 2 which... etc.
I do see what you're saying now. I see that 8 goes to ground and wondering if 2 and 6 should go to ground too but am out of my depth with this to be honest.
Hopefully someone who has been successful with this will post his/her connections to the unit and a close up of their boards front and back.
Calling all Gillespies; calling all Gillespies.
One other thing you might have already confirmed is that the outer portions of the repurposed rca jacks are going to ground and have a solid connection.
I'm not using the RCA jacks. I'm just inserting the module into the circuit that is at the center tabs on the volume control. And I doubt 2 and 6 go to ground because Dave's schematic would show their going to ground and he didn't do that.
I'm still trying to debug my Unity Gain module which, when installed, results in loud, buzzing distortion from the speakers. With the help of a multimeter or an oscilloscope maybe I can detect where in the circuitry the signal is getting so badly distorted. Does this make sense? Is there another method where I can use either of these tools to determine where the signal or DC flow is going bad?
If I were to trace DC voltage, one end is where the diode connects to the 5mA bias. The other end would be whatever comes out of pin 7 of the IC chip (and after the 50V cap and 100 ohm resistor). If I were to trace AC signal (i.e. I would have to trace an input signal of some kind), I'm thinking that would be before the .1, 50V cap that feeds chip pin 3 and after the pin 1 output that feeds the 100 ohm resistor.
What I'm after is where in the circuit the signal I'm feeding into it is becoming distorted. I'm thinking that as long as I can pick an entry point and an exit point in that circuit, I should be able to see at which point in the circuit (or which component in the circuit) the signal or the current is becoming "unwanted".
UPDATE: I watched some videos on how to debug a chip and surrounding circuit using an oscilloscope so I'm on my way ! Good ol' internet - always there when needed.
Thorn -- The connections on the schematic are correct. It is the connections between terminals 1&2 and 6&7 (which are connections external to the chip that need to be made) that cause the two op-amps on the chip to operate in unity gain mode (i.e. providing no gain or loss of the signal presented to the amps).
Question: If you disconnect the connections to the wipers of the volume control, does the loud buzzing noise go away?
Thanks, Dave, for the response. I'll have to open her up and reconnect the board. I'll post back.
Dave - If I connect only the GND and B- to the 500-C chassis, and then I touch the LEFT IN and RIGHT IN from the Unity Gain board to the center tabs of the volume control without disconnecting the existing leads to the center tabs, I get a reduced volume but no distortion. If I connect the LEFT OUT and RIGHT OUT from the board to the center tabs, I get the reduced volume AND THE DISTORTION.
I also found out that if I touch either of the IN leads from the board to either center volume tab (again without removing the existing leads on the center tabs), and I touch with my hand either of the OUT leads from the board, I get the distortion. It sounds like a 60 Hz ground hum.
Did you want me to do these tests with the existing leads removed from the center tabs? Thorne
UPDATE: Some other interesting facts... I traced the B- through the Unity Gain board. B- coming into the board reads -22 VDC. That goes into pin 4 of the LF353N chip. Following along, on the other side of the 47K resistor, the reading is -10.5VDC. That voltage goes into the 1M resistor, which goes into pin 3 of the chip, at which point reads -1.5VDC. Following the -10.5VDC, that goes into another 1M that goes into pin 5 and at pin 5 reads -1.5VDC. What's interesting is that pin 7 sometimes reads 0.6VDC and at other times it reads 0V. But when it reads 0.6VDC, the other side of the .22uF 50V cap always reads 0VDC. Does that make sense? The same test point reads between 2.7mVAC and 3.1mVAC. Same readings on the other channel.
I wonder if op-amp is damaged in soldering or by static discharge. I used socket from Mouser #575-293308.
Makes it easy to swap in the future too
Separate names with a comma.