Tube Amps / Music Electronics
|For current discussions, please visit Music Electronics Forum.||New: visit Schematic Hell!
The sunn still shines online!
|11/22/1999 5:54 PM|
||Can caps, star grounding, and hum|
I have constructed an amp using the Marshall 1987 schematic on an old Stromberg Carlson PA amp chassis. As it is there is massive hum, most of it originating in the first preamp tube. With the first tube removed there is just a moderate hum but with it in, when the volume is turned up at all, especially on the bright channel is makes an awful hum-from-hell sound. The heater wires are routed well and twisted so i'm thinking it's not from there. What I'm guessing is that i'm amplifying some noise coming up the ground circuit. So.......
I went to Randall Aiken's page and read about star grounding and compared it to what I have. I'm using the old (but still good I think) chassis mounted aluminum can electrolytics. It seems clear that all of this is using the chassis as a ground bus, which seems to be a pretty bad idea. So, the questions are:
Can I use star grounding but still use maybe the first can electrolytic and only seperate the grounds of the preamp stages and expect good results?
Is this level of noise consistent with a dirty ground line or should I be looking at something else?
|11/22/1999 6:23 PM|
although the can caps are connected to the chassis
I still connect a wire to them. In fact what I
do is connect the cathode of the power tubes,
and the CT of the PT, directly to the cap can.
Then run a big wire from there to my star ground
(which is actually another big wire).
|11/22/1999 6:41 PM|
But should I use the caps themselves? Clearly their ground connection isn't to an isolated ground star point but to the chassis itself. For the preamp it seems clear that this isn't the best idea but for the power tubes could you use them anyway?
|11/22/1999 6:48 PM|
Bear in mind that the hum level of the stock amp was probably at least reasonable when the amp was new and working correctly. Bad grounding layout would be the first thing I would suspect with a new prototype amp that was built from scratch. But in my experience, when a commercial product develops a hum problem, it's usually something else. I'm assuming that when you modified the amp, you didn't make any radical changes to the grounding arrangement.
The first thing I would do is check the first preamp tube by substituting a new good one. Preamp tubes can develop heater-cathode leakage, which will cause a great deal of hum, especially when they are several years or decades old. Next, I would make sure that the heater circuit is balanced with regard to ground. One quick way to do this is to set your voltmeter to AC volts and read between each side of the heater circuit and the chassis. You should see roughly equal voltages (on the order of 3.2 volts) on each side. Some amps grounded one side of the heater, in which case you should read 0V on one side and 6.3V on the other. This arrangement is noisier than balanced heaters, and you might want to consider converting to a balanced circuit. If you get a reading that floats around or otherwise just doesn't seem to make sense, then it's a good bet that your heater circuit is floating. A bad "humdinger" (hum balance) pot is often the cause of this. Floating heaters will cause a great deal of hum. When I was refurbishing my old Ampeg Rocket, I was amazed to discover that the stock circuit left the heaters floating. (It was right there on the schematic). Balancing the heaters with the typical dual 100-ohm resistor arrangement reduced the hum by a few tens of dB.
You can also check your heater circuit via resistance measurements, but the results might be a little misleading because of the low DC resistance of the heater winding and the tube filaments themselves. Voltage reasings are a bit less ambiguous.
Lastly, I would check the filter caps. I like to do this by AC-coupling my scope and looking for ripple on the supplies. A fair amount of ripple on the main B+ (that feeds the output tubes) is normal, but there should be virtually no ripple on the preamp supplies. It's important to do this both at idle and with the amp passing signal into a dummy load. If the ripple increases substantially as soon as you start turning up the amp, chances are the filter caps need replacement.
Once the simple causes are eliminated, then I would start looking at the ground layout...
|11/22/1999 6:58 PM|
Thanks, As it is, the tubes are new and swapping didn't seem to make a difference. The heater circuit's center tap is tied to chassis ground (with everything else) and I confirmed that each leg is 3.15 volts AC to ground. If the filter caps could be noisy I'll have a look at them and replace them. they are original (1960). While I'm at it, and getting rid of the aluminum can, chassis mount, many in one filter caps, I'll at least partially move toward the star grounding scheme. As for the original grounding scheme, I'm fairly close but the original didn't have this type of gain either.
|11/22/1999 7:17 PM|
Star grounding works, but is a pain to implement. I prefer to use a big copper ground buss. You can buy #8 solid copper wire at the hardware store that works great for this purpose. Mount a straight length of this wire across your chassis using insulated standoffs. Connect all your grounds to this wire "in order", meaning, the ground wire of your AC cord at one end, the preamp grounds at the other, and the rest in between in the order in which they appear on the schematic. Make sure that the B+ winding centertap and the ground side of the first filter cap connect to the same point on the buss, and likewise ensure that the ground side of the other filter caps are connected to the same buss points as the signal grounds for the stages they supply.
Connect the buss to the chassis at one point only, near the input jack. The best thing would be to actually solder a heavy stranded wire (say, #12) from the buss to the chassis right by the input jack, and connect the ground contact of the input jack to the same point with a very short piece of the heaviest stranded wire that will fit through the terminal. If you chose to use the input jack itself as a grounding point, use tooth washers and make sure the jack is nice and tight, and use only a good quality jack like Switchcraft. I only do that myself when soldering to the chassis is not an option (i.e., when using an aluminum chassis).
Buss grounding results in a much neater, more aesthetically pleasing and less cluttered layout than star grounding, and works just as well or better when done correctly. The key is to use the biggest, lowest resistance buss material you can manage, and allow no exceptions to the "all grounds to the buss" rule. Good luck.
|11/23/1999 2:32 AM|
You may have done this already, but do you have shielded wire going from the input jack(s) to the first preamp tube?
|Page 1 of 4||Next>||Last Page>>|