Hollis Ultra Flanger debug

[Yazoo]

I am trying to debug my build of the John Hollis Ultra Flanger. I get clean signal but no flanging. I have read the other posts. My problem is, I think, I am getting the wrong voltages on U1b and pins  7 and 8 on U2 the MN 3007 chip. I am getting the bias voltage of @ 4.9 volts but I am getting the following on U1b

Pin 8 9 volts
Pin 7 8.2 volts
Pin 6  6.3 volts
Pin 5  7.8 volts

U2

Pin 8 8.5 volts
Pin 7 8.5 volts
Pin 6  4.4 volts

From what I have read, the voltages for u2 pins 7 and 8 should be around  4.5 volts. Is this correct?

I have checked the board and I can't see any solder bridges. I am also getting 8.2 volts on the regen pot pin 1.

Any advice would be much appreciated.

[Yazoo]

I've spent the weekend trying to debug this but I haven't managed. I've double-checked for shorts but I haven't found any.

Here is the full list of voltages I am getting from a 9.3 volt supply:

U1
1  5.1
2  5.34
3  4.8
4  0
5  8.12
6  6.52
7  8.52
8 9.26

U2
1  9.24
2  3.6
3  5.1
4  0
5  0
6  4.17
7  8.86
8  8.86

U3

1  8.32
2  4.18
3  3.67
4  4.18
5  3.67
6  4.2
7  3.67
8  0
9  4.17
10  3.6
11 4.16
12 3.62
13  0
14  4.16
15  3.62
16  0

U4

1  0
2  3.7
3  4.2
4 4.2
5  0
6  1.8
7  1.74
8  0
9  2.16
10  1.37
11  1
12  5.75
13  0
14  8.35
15  0
16  8.35

U5
1  4
2  4
3  3.6
4  0
5  4
6  4
7  3.4
8 9.28

I have traced this though using a signal generator and an oscilloscope and I have tried an audio probe as well.

I think I can see a flange effect, what looks like several overlapping sine waves, on R13, output from U1b pin 7, but this is very low volume. Is this signal meant to mix with the straight signal? On the Geo build guide it says I should see about 4.9 volts on U1b pin7 but I am getting around 8.5 volts. I do get 4.9 volts from the R15/R6 voltage divider.

Any ideas would be very welcome. I hate having to give up on builds though sometimes I have had to.

[Yazoo]

I think the problem with my build lies between U1b and U2b. The voltages are just not right. I am trying to understand what is going on with the voltage going into U2. From the schematic,  I can see 9 volts is going into U2 pin 1. It then goes through a 47K resistor to U2 pins 8 and 7 and then through 2 more resistors and a capacitor to the Vb supply of 4.9 volts. So are the three resistors acting as a voltage divider? If you have the time and the inclination could you have a look please? The schematic is available here http://www.hollis.co.uk/john/ultraflanger3.jpg

[anotherjim]

I think you should be getting about the same voltage out of the (IC2) BBD chips pins 7 & 8 as is entering it's pin 3 (5.1v). Is the 47k "pull up"  resistor to + supply there good? If you disconnect the other 47k to the next amp (IC1b), it's output should swing back to the reference voltage (4.9v) and see if the IC2 pins7/8 drops to closer to pin3 volts.
Point is - see if the high voltage at IC2 pins 7/8 is due to IC2 fault or external influence (back from IC1b).
If you disconnect the 47k pull up resistor, and there is still nearly 9volts on IC2 7/8, I'd suspect a bad BBD chip. It shouldn't put out any voltage at either pin 7 or 8 without the pull up resistor.

[Yazoo]

Thanks for your suggestions. I'd got to the point where I was starting to remove components. It never occurred to me that the chip might be faulty. I've got another one ordered and I'll try out your suggestions.

[Yazoo]

I've lifted the pull up resistor, R7 and I am getting @ 1.2 volts on pins 7 and 8 of the MN3007. On U1b I am now getting 1.3 volts on pin 5, 5.4 on pin 6 and 6 on pin 7. Does that look like a bad MN3007?

[anotherjim]

It's getting hard to help you with this. You've linked a hand drawn schematic but are obviously working to one with the parts numbered. I don't know those numbers!
From what you just posted it does look like it's a faulty 3007. IC1b output voltage looks a bit suspicious, but that doesn't explain the high output of the BBD, which is the first problem to solve.

Anyway, the outputs of the BBD are from "source follower" P channel mosfets, they only have a ground connection to their Drain in the chip. They need a positive voltage supplied via the pull up resistor to work. When working the BBD is moving the voltage at the input to the outputs through it's delay line. A source follower has no gain, so a very similar voltage to the input should make it all the way to the output -  but only if the pull up resistor is there.
So, it does look bad for the chip and a lesson to teach you to have a spare! That's always a good idea with any of the parts in a circuit, but more so with complex parts like an IC.

[Yazoo]

Thanks for your patience. I do appreciate your help. I've got another MN3007 on the way. I assumed I had made a mistake so I've spent hours poring over my pcb with an eyeglass and a pin, making sure there were no bridges or mistakes.

The numbered layout I was referring to is on Geofex: http://www.geofex.com/pcb_layouts/layouts/ultrafln.pdf

[anotherjim]

That's better!
I see now you have a bias voltage trimmer. The BBD signal will be louder and cleaner with that correctly set to suit the BBD. You can't fix a precise voltage at which this is best because it varies from chip to chip, but I don't think the bias setting can throw the BBD output out by as much as it is.
You could remove the 3007 BBD and put a temporary wire between it's pin 3 and either pin 7 or 8 (make sure R7 is disconnected). That would simulate a working BBD as far as the DC voltages go. So then you should be able to measure the "approx 4.9" bias voltage where the diagram shows.

[Yazoo]

Great thanks. My voltages are looking a lot better. I'll try the new chip when it arrives so fingers crossed!

[Yazoo]

The new MN3007 arrived today and it is now working correctly. Thanks again for your help.

On to the next build!