Tube Amps / Music Electronics
|For current discussions, please visit Music Electronics Forum.||The sunn still shines online!|
|4/4/1998 7:43 PM|
I've recently built the guitar preamp from O'Connor's Tonnes of Tone book and, while it sounds good, it has a bad hum/buzz. It does this even w/ the volumes off. I'm using a Hammond 270AX PT and I'm running the heater's off of 6vac, not dc. For simplicity's sake, I'm trying to avoid doing that. I've referenced the heater wires to ground w/ a couple of 100 ohm resistors w/ no change. I'm guessing that I have a ground loop or some sort of grounding problem. It kind of sounds like when you unplug the cord from your guitar and touch the plug w/ your finger. I'm not knowledgable enough to start poking around so I'm trying to get some input. Help!
|4/6/1998 7:22 AM|
Make sure that your preamp grounds are tied together at a single point and a wire from that point connected to your ground. There is a lot of gain in that circuit and even a tiny bit of resistance to ground at certain points can result in hum.
One idea is to isolate the input jack from the chassis and connect a wire from the ground side of the jack directly back to the power supply ground. Be sure to use shielded wire for the input and at any other high impedance connections in the amp. Ground the shielded wire at the jack end only.
If that fails, a balance pot on the heater secondary may be required instead of the 2 x 100 ohm resistors.
|4/6/1998 11:46 AM|
The whole grounding thing confuses me. Is there supposed to be more than one chassis connection? I've got the power cord connected. Other than that, the only connection to chassis is the input jack by virtue of being in contact w/ the chassis, I suppose. The power supply ground you speak of...Is that where the CT wire of the PT is connected?
As for the shielded wire, right now I have the shielded wire grounded at both the input jack and the tube end, as per O'Connor's drawing. If I only have it connected at the jack, do I also need to isolate the jack? I've tried disconnecting the grnd from the tube end, w/o isolating the jack and then measuring resistances to chassis, and they measured something like .895m. If I leave it connected, they measured .6 k(I think). Or maybe it was .6 ohms? Anyway, the difference was pretty extreme so I was afraid to power it up.
Sorry for any bad terminology. The only reason I tried this project was because he had the layout drawing, so I felt a little more comfortable. Otherwise, I'm a bit lost.
|4/6/1998 12:49 PM|
Ground can be:
- a zero voltage reference point
- a power supply return
- an electrostatic shield
Properly, the chassis should be an electrostatic shield, not a power supply return or a zero reference voltage point.
A zero voltage reference point can only be zero volts if there is negligibly small current flowing, given that it has to be made from real conductors.
To avoid getting either feedback because the power supply currents cause voltages across the chassis resistance, or hum, you should only connect to the chassis at one point. This guarantees that no current flows through the chassis, and can't be a source of hum or feedback. This includes the ground of the input jack, as this ground is taken by the input stage as common mode signal.
It is *possible* to use the chassis as a power return/signal ground and get away with it. Fender does this, but they also spent a LOT of effort on tinkering and cut-and-try. In general you can't guarantee success this way.
There should be a single point where the power line safety ground wire, the power supply (-) return, and the signal ground are connected.
This is a version of star grounding.
For lowest hum, you should connect the power transformer CT to the filter capacitors, and then run a separate wire from the filter cap (-) to the single ground point. This keeps the voltages generated by the power supply filter charging pulses out of the system ground, as only pulses flow in the CT, and only DC flows in the power return line.
The power return should connect to the star ground point, which is the single connection to the chassis. As a practical matter, the signal grounds and power returns for various sections are grouped locally and wired to the star ground point to cut down on the number of wires going to the SGP. The signal grounds to the input jacks, tone/volume contros, etc. should also be brought to the SGP, not tied to the chassis, especially the input jack, as any interference is amplified just as the signal is.
If you run shielded wire to the input jack, insulate the jack from the chassis, and let the shield "ground" the ring contact on the jack, also "grounding" the shield at the tube end.
|4/6/1998 6:46 PM|
Okay, I think I understand that. How would I go about isolating the jack from the chassis w/o buying a special jack?
As an experiment, I removed the input jack from it's hole, sort of isolating it I guess, and it didn't help the hum. Still might not be a bad idea to isolate (insulate,whatever) it anyways.
How about this. I measure the voltage on the filament wires and I've got 7.28vac. Would that be a possible reason why I have such hum?
Also, I used 1/4 watt 100 ohm resistors to reference to ground with as I didn't have any 1/2 or 1 watts handy. Would these still work? At least temporarily? I just used the 1/4 watts to see if it would help, w/ the intention of getting higher rated ones sometime soon. Which is recommended?
Okay, now I'm gonna go soak this thing in gasoline and see if that helps....
|4/7/1998 2:08 AM|
Okay, I think I understand that. How would I go about
isolating the jack from the chassis w/o buying a special
Seems like there are some Rean jacks in the Mouser and Parts Express catalogs that have plastic bodies. I've not seen these puppies in real life but I intend to order a few since it looks like they would isolate the jack from the chassis.
|4/7/1998 2:06 PM|
Switchcraft makes a nylon bushing version of its standard 112 jack. Mouser sells it for about $0.30 more than the metal bushing version.
|Page 1 of 4||Next>||Last Page>>|