| cantskienuf |
| I built the BrianGT LM3875 premium NIGC kit and think it sounds great; however, I have a turn-off thump. I built the units as true stand-alone monoblocks with individual 22-0-22 Avel 160VA toroids per side. I used a 0.1uF, 450v Auricap in the Cz position and a 2.7 ohm 5 watt Kiwame in Rz. I added a 56K Riken shunt on the input and a .22R Mills on the output ala Decimal Dungeon. Other than that, it is stock. I have no hum or buzz, just a thump when power is turned off. It is not particularly obnoxious, but I'd like to get rid of it. Any ideas? |
|
|
| Tazzy |
| Power switched relays on the speaker outputs :rolleyes: |
|
|
| Peter Daniel |
Properly implemented kit will not cause turn off/on thump.
It must be all those extras ;) Seriously talking though, check your grounding, this usually creates that kind of problems.
Power switched relays on the speaker outputs will help, but will not fix the problem. |
|
|
| Sandy H. |
Does it happen if you leave the input (CD player etc) turned on? Does the same thing happen if the input is turned off first?
I had an amp which had both turn-on and turn-off thump unless the source was powered on first. I'm not sure what the real fix was, as I was fine with just turning the CDP on first and off last.
Sandy. |
|
|
| Peter Daniel |
| Also, what's your DC offset at the output? |
|
|
| rnoble |
| I would take your extras out of circuit and test. I have done the same setup recently, (minus the zobel and extras) with 2 225VA transformers and it's dead quiet on startup. |
|
|
| sek |
| quote: | Originally posted by Sandy H.
I was fine with just turning the CDP on first and off last. |
Not to offend anyone. :D
But isn't this the ALWAYS recommended way of switching ANY chain of audio devices that don't ALL feature a turn-on/-off mute?
I even learned to switch on the computer screen on last and off first, back in school! :att'n:
:bigeyes:
Sebastian. ;) |
|
|
| cantskienuf |
| quote: | | Also, what's your DC offset at the output? |
Peter, I haven't tested DC offset lately. The last time I did they were both at approximately 40mV into an 8 ohm load. Brian thought that sounded OK, but I've seen from other posts that some are able to work the offset down much lower. Do you think that is the culprit? The thump has always been there, but I think I'll re-check the DC offset tonight and make sure it hasn't changed significantly.
| quote: | | and it's dead quiet on startup |
rnoble, mine are quiet on startup also. It is the power-off thump that I want to work on.
| quote: | | Does it happen if you leave the input (CD player etc) turned on? Does the same thing happen if the input is turned off first? |
Sandy, I always turn on the isolation transformer first, then the pre-amp, wait for warm-up, then power up the CDP, then power up the amps, then press play on the CDP, then bring up the attenuators. For turn-off I first turn down the attenuators, then press Stop on the CDP, then power off the amps, then power off the pre-amp, then power off the CDP. I don't like to hear what happens in the pre-amp (Foreplay) when the tubes warm up, or shut down. |
|
|
| Peter Daniel |
| 40mV is fine, but maybe your source components pass DC as well so check with a source connected. |
|
|
| sek |
Hi,
| quote: | | I've seen from other posts that some are able to work the offset down much lower. |
40mV is not extremely much, but it depends on what the speakers make of it, though.
Do you have measurement equipment like an oscilloscope to verify what happens at turn off? I guess, not.
Another approach: have you tested the GCs with the input devices disconnected?
For further guessing, photos of the amp internals would be of great help...
Sebastian. ;)
edit: Hehe, Peter was quicker, but pictures would still be great. ;) |
|
|
| cantskienuf |
Peter & Sebastian,
You are right that I previously checked DC offset with no source attached. I don't have an oscilloscope, but I'll see what I get from the DMM tonight with both a source and a load connected and in the state present when I typically turn off the amps. I'll try to get some pictures, but my digital camera never seems to show the detail necessary to adequately display my wiring.
Thanks for the help so far! |
|
|
| cantskienuf |
OK, did some measuring last night and found that I have approximately 70mV of DC offset into 8 ohms with the pre-amp and CDP connected and powered on. Attached are my wiring layout and a couple of pictures.
All AC is isolated on one side of a 1"x4"x7" aluminum heat sink. Positive and negative rails pass from the PSU board through individual 1/4" holes in the heat sink. Ground passes back through a separate 1/4" hole. Everything is mounted on a single top plate. Binding posts and RCA are isolated and connected only to the chip board. Maybe this will help you to see if I erred in my grounding scheme. |
|
|
| Peter Daniel |
As it seems from your diagram, the ground pin from AC connector is attached to the top metal panel. It may be the case that this earth pin is not really grounded (with your house wiring).
Another posibility is that having wooden chassis, you might not have enough shielding.
For me, whenever I had problem with that sort of noise (light switches making noise in speakers when switched), I always connected the grounded binding post (from output to speaker) directly to the chassis and disconnected earth wir from the chassis.
In case of monoblocks it is not an issue, but in case of stereo amp, both channel posts (grounded) have to be connected to the chassis, preferrably at one point. |
|
|
| Nuuk |
The 'extras' don't cause any such problem with any of the GC's that I have built.
My guess (and that's what it is) would be that the thump may be due to having the transformer close to the amp circuit. When power is removed, the magnetic field collapses and there is still enough 'juice' in the PS caps to allow the amp to make the thump. :att'n: |
|
|
| Peter Daniel |
| I've seen many GC amps with transformers basically on the board and no thumps reported. |
|
|
| cantskienuf |
| quote: | | As it seems from your diagram, the ground pin from AC connector is attached to the top metal panel. It may be the case that this earth pin is not really grounded (with your house wiring). |
The AC safety ground is 12AWG copper wire soldered to a copper washer at the central grounding pin. The amp chassis ground wire is 21AWG copper in teflon soldered to the other side of the copper washer. The washer is then bolted to the top plate. I'll re-check the solder connections, the tightness of the bolt and continuity from amp chassis ground to the IEC connection to see.
I'm beginning to wonder if it could possibly be a property of the on/off switch I used. It has a pretty solid throw, and I usually give it a very definite move when turning on or off. I think I'll do some testing on it this evening to see if it is breaking the circuit smoothly. It is a simple 15A SPST from Partsexpress. Ever found this to be a problem?http://www.partsexpress.com/pe/psho...tnumber=060-371 |
|
|
| cantskienuf |
| Found a solution, but not the problem. Of course it wasn't the switch. The grounds showed good continuity throughout the circuit also. However, after much experimentation I've found that if I turn them off while actively playing music, there is no thump. As you'll note from an earlier post, I normally turn off the power amps first after having reduced my pre-amp attenuators to their lowest setting, which is effectively mute. If I leave the CPD actively playing music and reduce the attenuators to their lowest playing setting, rather than to mute, then turn off the power amps, they power down smoothly with no unwanted noises. It still leaves me thinking that something is wrong in my application of the kit, since I have to trick the amps into starving themselves at power down instead of being able to exit from a steady state without an active signal. It seems that something is going on in my amps when they are not busy amplifying music. |
|
|
| Tim__x |
| Measure your DC offset with no source and your attenuator at it's lowest volume setting. I bet it'll be a fair bit higher than 70mv. |
|
|
| kneadle |
Just so I know I'm doing this right: how do you measure DC offset with a DMM?
I just finished assembling my BrianGT, and I'm trying to avoid the above described problem. So I stuck my DMM in either pole of the output binding posts; nothing is connected to the amps ins or outs. One channel reads about 99mV, the other is at 72mV. I'm hoping that's acceptable before I hook a source and some speakers up.
I'd also like to learn to troubleshoot if I experience similar problems described in this thread.
Thanks,
Dave |
|
|
| ransom peek |
Isn't it likely that the thump is simply due to the relatively high power of the transformers, (160VA x 2) totalling 320VA?
Has the thump problem been resolved? |
|
|
| cantskienuf |
| I'm not sure if it is resolved or not. I've moved the amps to a different system and it is not as noticeable. I read a thread on another forum that was suggesting a capacitor bridge on the switch (.01uF 600V) of a tube amp to eliminate what was happening when the transformer field collapsed at turn-off. I've tried that on one amp and indeed seem to have a favorable response, but I only get minimal turn-off sounds when using the amps in this system anyway. I'm thinking of moving them back up into the main system to see exactly what this approach does (more sensitive speakers). I don't know that I like this solution because it leaves about 2 AC volts in the amp when powered down. That is not really a problem for me because I use an upstream AC cutoff on all of my dedicated audio AC lines, but it doesn't seem to be something I would recommend. At this point I am suspecting that the close proximity of the transformer to the switch, fuse and AC lines is probably the source of the thump. I haven't had an opportunity to try re-routing wires at differing angles to see if I can minimize it because I'm working on a pair of 2A3 SEs. |
|
|
|
