I bought this DBX 160A from a touring company. The comp was made in 2004 in China, it’s 15 years old, so I decided to recap it besides modifying. I’ve modified the 160A before, two later units from 2010 I believe. They contained 105° caps and the THAT 2181 VCA (marked as DBX V1B, but it’s obvious what it’s really is because of the resistor configuration). This DBX was full of Yageo 85° caps and THAT 215x VCA (DBX V1), so not every Chinese DBX 160A was born equal. And it was just beginning.

Harman guys don’t want to provide a schematic for the latest 160 series edition, so there is my photo of the component side of the PCB as a replacement. You can download the original Full res photo of the DBX 160A PCB.

DBX 160A PCB Top

And the mirrored other side so you can read it side to side with the components. You can download it as well and make an overlay image in a photo editor with the previous one.

DBX 160A PCB Bottom Mirrored.jpg

But if you are as lazy as I am there is the overlay for you. It's also has a huge version - DBX 160A PCB Overlay.

DBX 160A PCB Overlay

As I mentioned before, I’ve modified the DBX 160A earlier so I skipped the tests and just start to desolder everything that I wanted to replace.

1. U1 -> socket. The 5532 family is not bad, but there are enough op-amps that are better in every way. This circuit is stable enough, it can handle even ultrafast chips.

2. U5 -> socket. The same reason.

3. C2 -> WIMA MKS2 10 uF (MKS2B051001N00KSSD). The original cap is the 22 uF bipolar electrolytic, but 10 uF is enough there. The phase shift at 20 Hz is about 6-8° which is a good result. I’ve tested Pana FR 33 uF in this position once and even this capacity didn’t give me the flat phase, I remember something about 3-4°. So if you are a phase purist you should go wild and put significantly larger caps there.

4. C3 -> WIMA FKP2 0.033 uF (FKP2C023301L00HSSD). Since there is the official place for the shunt I just used it the best way I could.

5. C35 and the other big can in the PSU -> Pana FC 2200 uF (EEUFC1V222S). The 3300 uF low profile version (EEUFK1V332S) fits, but it’s more rare and pricy. The 2200 uF caps were what I had at the moment on my table. I had to bend the leads though, anyway.

6. C40 and C41 -> Pana FR 560 uF (EEUFR1V56). Yep, they are big but it’s the closest to desirable what I had at that moment and I had a BIG stash of them.

7. Every other electrolytic c(r)ap -> the closest Pana FR.

8. L1, L2, L6, L5 -> Murata ferrites BL03RN2R1M1B. It’s not a very significant improvement, but I’ve tested them against the “blue” inductors from the later 160A (this unit contains orange ones). The ferrites are 0,1-0,2 dB louder, the measured noise was lower and they are worth almost nothing, so why not?

This is the photo of the marked PCB.

DBX 160A Marked board.jpg

That’s it. I didn’t swap the VCA just because I didn’t have them then but I will.

DBX 160A Mod

P.S. So… Do you remember about “not every Chinese DBX 160A was born equal”? Besides cheaper caps and the older VCA, this one has the ultimately crappy PCB. Desoldering was hard as hell. It’s not my skills or equipment. I’ve desoldered enough components in my life. I have the very good Hakko 474 desoldering station. And I’ve modified the other 160A without a problem. But in this one, every solder point tried to live its place and got nearby trace as a company after the slightest touch, so I had to be extremely accurate. Refilling joints with a soft lead solder helped very much. So be prepared!