• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Decware Zbox - looking for Tube buffer mods to reduce buzz

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Recently got a used Decware Zbox. Unit has a buzz in both channels. Checked out all the parts and all parts are good. The internal wiring is a rats nest and may windup redone. But before going there, looking for some basic "quiet" mods that might help. Some of the items I am thinking about

- two 100 ohm resistors to ground each leg of the heater power supply
- reversing the phasing of the heater supply
- adding a choke to power supply
- adding more capacitance to power supply
- converting the AC heater circuit to DC
- adding an OC2 tube to power supply

I'll post some schematics of the Zkit4, which looks very similar to the parts and layout I have been testing. So from a concept model, any mods to make this circuit quieter, should work on the Zbox (no schematic available).

The schematic with the OC2 is an upgrade that is not in place, but certainly worth considering.
 

Attachments

  • zkit4schematic.gif
    zkit4schematic.gif
    5.2 KB · Views: 499
  • zkit5schematica.gif
    zkit5schematica.gif
    6.9 KB · Views: 463
  • zkit5schematicb.gif
    zkit5schematicb.gif
    6.9 KB · Views: 473
  • zboxreva.JPG
    zboxreva.JPG
    117.5 KB · Views: 475
  • Picture 15.jpg
    Picture 15.jpg
    138.9 KB · Views: 476
Last edited:
first thing I would do is dc on the filament

I'm thinking that as well. I have a 2"X2" board that will regulate from 1.5 to 24 volts at 1.5 amps and can be adjusted to 10 volts which is what the AC heater voltage currently is. The heater supply looks like it's .3 amps so it's not problem for the board. Not sure if going ac to dc is one to one on voltage.
 
I would sort out the wiring first. I am astonished that such poor quality building is being commercially sold, even from China. Was it a kit? Mind you, the daft circuit design is perhaps a warning.

First step: identify the wires from the transformer to the rectifier bridge. Ensure that they follow the same route and are at least roughly twisted together. Keep them away from all other wires.

Step two: identify the wires from the rectifier to the first (reservoir) capacitor. Do the same for them.

As currently wired, the transformer wires (red?) look reasonably OK but the cap wires (grey/pink?) are appalling. They don't follow the same route, and it looks like the negative wire is twisted around a grounding wire. Awful! The ground wire should connect to the cap, not the rectifier, and be kept away from the rectifier-cap wires as much as possible except where it actually joins.

The aim is twofold: reduce the area of loops carrying charging pulses, and stop injecting buzz into the ground connection.

Make that change and report back. It is likely that the unit has other wiring problems too.

PS just noticed that the Decware website claims that the units are handmade in the USA. Do you have a genuine unit or a chinese clone?
 
Last edited:
The grid stoppers in the second photo do not look like 22k. They look more like 47K. Is that correct? Hard to tell, they may be 470R which is probably OK. I would think 1K should be sufficient. I agree with DF96. The PS wiring needs to be corrected. The wiring is abysmal. I'm surprised that work would come out of Decware, but I don't have any direct experience with their products. That circuit should work fine with an AC filament supply, but you would certainly want a tighter twist that what is shown, and routed more like the first photo. I also notice that the first photo in/outs have been changed to shielded which may help.
 
Last edited:
Thanks for the wiring walk thru..will give that a go first. My unit is a Decware made unit. I've seen a lot of photos of this box and they follow the rats nest lay out.

One of the thoughts I had, and verified, is that the hum/buzz would probably be less noticeable on the flea powered amps Decware sells. So I tried the Zbox on a 4 WPC SE tube amp and the hum/buzz was barely noticeable. I'm trying to use this box on a few tube amps that range from 50WPC to 350WPC. Seems like th higher up in power you go, the more noticeable the noise is. Same problem with SS amps . So maybe the craftsmanship is good enough for flea power amps. Certianly theres much to be improved. I go thru the wiring this weekend.
 
The schematic is totally wrong for the Deware Zbox. Gain is said to be close to unity and Zout ca 1k. According to the tech papers the Decware is a simple CF with a ECC83. Sound should be better without it as it can only add colouring. You then also get rid of the buzz;).

The Zkit is completely different as it is a gainstage with the highly unlinear ECC82.
 
well, I removed the pot from the housing, the tube socket from the chasis to see if any basic wire bundling could be done. Of course, as luck would have it, as soon as I moved the POT and the tube socket, a few of the solder joints popped loose. So now we have cold solder joints to add to the mix. The picture I just upoaded is a PCB for the Zkit4, which I think is the same circut as the Zbox. It would have been great if this unit had a nice PCB instead of the point to point wiring with what looks like extra long wires. This thing is screaming to be redone.
 

Attachments

  • zkit4sm-1.jpg
    zkit4sm-1.jpg
    14.3 KB · Views: 305
The schematic is totally wrong for the Deware Zbox. Gain is said to be close to unity and Zout ca 1k. According to the tech papers the Decware is a simple CF with a ECC83. Sound should be better without it as it can only add colouring. You then also get rid of the buzz;).

The Zkit is completely different as it is a gainstage with the highly unlinear ECC82.

I think the white papers is for current production. The Zbox ships with a 12AU7/ECC82. A lot of folks use 12BH7's. Using 12AX7's/ECC83's is reported to be problematic becuase of input over load. A POT is provided so that if you use tubes with other MU's , line level adjustments can be made. So basically, if you use a 12AT7 or 12BH7, there's some gain and may need to be rolled back. I found the box to be quieter using 6414's. The 6414 is a longer tube, and may be lest influenced by the effects of a starved AC heater as used in the Zbox. In any event, there's a lot of opportunity to put in a few mods as pointed out in this thread to quiet the unit down. I'm going to try them in order of level of effort. It would be great if I could come up with a measurements scheme so I can actually measure the effects of the changes. The buzz is pretty load, so maybe just a basic Iphone sound analyzer program. I dip into this over the next few days. more to follow.
 
Can we be clear: are we talking about the Zbox (cathode follower, gain<1, low o/p Z) or Zkit (degenerated grounded cathode, gain>1, high o/p Z)? Two very different circuits with different characteristics, although perhaps in similar boxes.

There is nothing wrong with p2p wiring, provided it is done properly. At least you can change it easily; a bad PCB design would mean starting from scratch.

If the buzz is loud then you don't need any instrumentation to show an improvement. All you need to do at first is sort out the PSU wiring, and ensure the heater supply has a voltage reference rather than being left to float. Then send the bill for your time to Decware!
 
OK, I'm going to attribute this to 4 cold solder joints and poor wiring, which I would have not found if you guys did not talk me into doing something about the wiring. When I moved the POT and the tube socket, wires started to pop off their joints, so at least 4 cold solder joints where found. While in there, I move the tube socket 45 degrees so that the AC heater lines were not in the middle of the tube circuit components. I moved the turns on the ac heater lines as far forward to the tube as I could and hot glued the rest of the AC heater lines right to the box case. I positioned the B+ pink wire as far as I could from both the in and out lines and the lines going to the pot. I tied the in and out lines together and made sure they were not near AC or DC lines. Tested it out and there's no hum or buzz. Thanks for the input , I was going to dive right into a AC to DC conversion which would not have addressed the real issue. I decided not to do the two 100 ohm resister mod to the heater supply. The current loss would have been .1, I think the power supply can only handle .3 and the 12AU7 tube needs .3 for heater current so thats out. Thanks again everyone.
 

Attachments

  • photo-1.JPG
    photo-1.JPG
    558.7 KB · Views: 323
  • photo-2.JPG
    photo-2.JPG
    596.1 KB · Views: 308
  • photo-3.JPG
    photo-3.JPG
    617.1 KB · Views: 307
Last edited:
Does the heater line have a DC reference? It should have. Ground is often sufficient - just ground the heater CT (valve pin 9). Don't do this if one side of the heater is already grounded.

I'm buying you a beer. There was a faint background hum remaining but only noticeable a full volume with no signal. I checked pin 9 and it's not grounded. I confirmed that the whole AC heater circut is floating with no grounds. I grounded pin 9 and the system when totally dead quiet. No noise at all. Awesome. Thanks.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.