GAStan wrote: ↑Tue Oct 10, 2023 2:29 pm
Right now the biggest issue I have is with microphonics. I tested my tubes in another amp and all are at least a little microphonic when tapped, some very little, some ring like a glass bell, the rest somewhere in between. The seasoned JJ tubes I have are the best I have.
All sockets are Belton Micalex.
The JJ's, when installed in V1 of the ODS, give a slight low pitched "thunk" when tapped. The resistors and wiring to the grids are very sensitive to chopsticking. CL2 is most sensitive. I removed the 500pf ceramic capacitor with no change. Are CF resistors known to be microphonic? The orange drop cap can be heard when tapping it, but at a much lower level as if it is just the vibrations transferring through the leads. This cap is siliconed to the chassis. CL2 exhibits this behaviour with the CL Volume at zero. CL1 is similar to CL2 when CLV is turned up. All of the above is with OD off.
Moving to V2 the OD entrance trimmer was turned to zero basically grounding the OD input. Grids are microphonic but not nearly to the extent of V1. V2 plates are microphonic and tapping the plate wire can be plainly heard.
V3 is good, even with a tube in it that rang like a bell when tested in the other amp.
All power tubes are good as are the components.
I'm at a loss, what should I check next and how do I check it?
In my experience, the whole of V1 is super sensitive and will definitely be microphonic no matter what you do. Even touching the V1 grid wire with a chopstick will produce a sound. I am unsure what the level is of microphonics you are experiencing though and whether this is an indication of some other problem. I also noticed that your hookup wire is stranded. I always use solid hookup wire (as did HAD) so not sure if that is adding to the microphonics?
The bigger question would be, are the microphonics negatively impacting the sound of the amp? Can you hear artifacts while playing or is it only when you chopstick? Without having the amp in front of me and hearing/seeing the issue it is difficult to diagnose.
Thanks Ian for this post.
Great question and point! Hard to say if it has a negative impact. So far I've only been able to play it at bedroom levels except for one time and that time we didn't notice anything but weren't specifically listening for them either.
The wire I used is overcoat, stranded but fully tinned like when you tin the end of a wire prepping it to solder to a termination. It bends and stays in place similar to solid. When searching the forum for microphonics I did come across a thread where stranded was actually used to replace solid to fix a microphonics issue in a TW style build. I'm not sure what to think...
It is worse when tapping the grid resistor or feedback resistors than the wires themselves on V1.
I have some more tubes on the way from TubeDepot.com that have specifically been tested for noise and microphonics so I'll just wait for them and hope they're better than what I have. In the meantime I'll use the JJ's I have and move on to tuning the PI, OD entrance and FET levels after I put the 500pf cap back in.
Had no problem with the procedure, it's pretty straightforward. Problem is I ran out of adjustment before reaching zero mV. So I took some other measurements. After removing the cathode jumpers I checked the bias and the tubes vary by 3-4 mA. Been thinking this over this morning and concluded I do not know if the variance is the tubes themselves or the position/slot they are in, or very possibly a combination of the two now that it has several hours burn-in time. So this evening I'll rotate each tube through each position and document its mA reading without changing the bias. Then I'll select tube position based on the documented readings and hopefully I will get a more balanced reading across the four tubes.
The PI I have installed now is the one I'd like to keep if possible. While it rings like a bell if tapped in V1 it is actually very quite otherwise, the quietest tube I have. It is not at all microphonic in V3. It's a Mullard new production ECC83 from TubeDepot with their balanced triode option. I don't have a test set so cannot verify matched triodes, have to trust them.
After reading several discussions of what different members thought HAD's goal was as far as which section created distortion, I have formed my own goal for this build. I want to keep the power section as clean as possible and let the OD section of the preamp create the distortion. The reason is simple-low volume is needed most of the time in our small house. However my son, who this amp is ultimately intended for, at 14 is already in a rock band with his friends and does have the opportunity and need to crank it up during their practices.
Hi Glenn, how did you do the grounding of the pot housings? If there is a thick layer of powdercoating on your chassis, there is a chance your pots are not grounded. And if your coax is soldered to the pot housings… This could be the reason of your noise problem.
gonzomode wrote: ↑Tue Oct 17, 2023 9:59 pm
Hi Glenn, how did you do the grounding of the pot housings? If there is a thick layer of powdercoating on your chassis, there is a chance your pots are not grounded. And if your coax is soldered to the pot housings… This could be the reason of your noise problem.
Great point!
I used grounding washers, the type with the internal tabs that are raised/twisted to cut through paint and powder coating. I checked continuity for each pot as I installed them and several times since. They have checked good each time.
Right now I'm working with AES about the 6L6 tubes I purchased from them, Apex matched Sovtek 6L6WXT+. When I initially installed this set the all biased at the same mA reading, I set them to 30mA. After about 10 hours burn in time they have drifted apart, from 28 to 32 mA now. I did rotate the tubes through each position here are the results.
20231016_190849.jpg
Apex tube ID down left side, position across top. Green dot indicates where I installed each tube, red and green are subsequent readings. As shown position did not have much effect as each tube was pretty consistent in each position.
The variance in tubes has resulted in one side requiring the most mismatched tubes in order to balance the two sides of the OT.
Personally I think the Apex matching is a great concept and implementation-to a point. However based on my experience with this set of tubes I believe they would benefit by burning in ALL of their tubes prior to matching them. They do offer a burn in service but only on select tubes, primarily JJ's. Unfortunately they do not offer burned in Sovtek 6L6's, I did inquire prior to purchasing these.
You do not have the required permissions to view the files attached to this post.
Fwiw 2 cents ….this is the reason I won’t buy tubes unless they are burned in … modern production need those plates hit with actual high voltage for a period to become stable . Too many tubes die as soon as you hit them with an actual high voltage . I have sent back 2 in the last few months out of quads . One was burned in . Died as soon as it was installed and hit an open chord . Not a good success rate considering I don’t buy tubes in bulk . Also, a 4ma drift over 24 hours is bull shite. Yes, unfortunately JJs usually only get that service through AES.
GAStan wrote: ↑Wed Oct 18, 2023 2:25 pmThe variance in tubes has resulted in one side requiring the most mismatched tubes in order to balance the two sides of the OT.
You are within 1 mA side to side now? That's more than close enough for the idle current. Gm is unknown, however.
You might want to try Eurotubes: "After a 24 hour burn in all power tubes are tested for noise and matched at multiple operating points for plate current (± one milliamp) and transconductance."
General comment: Tubes should be burned in *and then* matched. Not the other way around.
To match, and to let them drift afterwards while they burn in is utterly nonsensical counterproductive, and wasteful. Some vendors burn their tubes in before matching, some do not.
Others do not burn in and only match for bias current, and do not measure for gm. Its a wild world out there.
GAStan wrote: ↑Wed Oct 18, 2023 2:25 pmThe variance in tubes has resulted in one side requiring the most mismatched tubes in order to balance the two sides of the OT.
You are within 1 mA side to side now? That's more than close enough for the idle current. Gm is unknown, however.
You might want to try Eurotubes: "After a 24 hour burn in all power tubes are tested for noise and matched at multiple operating points for plate current (± one milliamp) and transconductance."
Thanks for the recommendation Martin. Looks like the only carry JJ's, but at the best prices I have seen. I really want to try a well matched set of the Sovtek tubes so ordered some from TubeDepot. After a good email exchange with Toby I feel comfortable giving them a try. He said their process is Match, Burn In, Re-test. I have ordered pre-amp tubes from them with pretty good results. The one I received that was noisy was quickly exchanged for a better one-the whole return process was actually kind of pleasant working with Michelle.
jelle wrote: ↑Wed Oct 18, 2023 8:44 pm
General comment: Tubes should be burned in *and then* matched. Not the other way around.
To match, and to let them drift afterwards while they burn in is utterly nonsensical counterproductive, and wasteful. Some vendors burn their tubes in before matching, some do not.
Others do not burn in and only match for bias current, and do not measure for gm. Its a wild world out there.
I am finding this to be true.
I wonder if we could have a thread or even whole section dedicated to sharing our experiences, good and bad, with various venders.
jelle wrote: ↑Wed Oct 18, 2023 8:44 pm
General comment: Tubes should be burned in *and then* matched. Not the other way around.
To match, and to let them drift afterwards while they burn in is utterly nonsensical counterproductive, and wasteful. Some vendors burn their tubes in before matching, some do not.
Others do not burn in and only match for bias current, and do not measure for gm. Its a wild world out there.
I am finding this to be true.
I wonder if we could have a thread or even whole section dedicated to sharing our experiences, good and bad, with various venders.
And then share it with the vendors and throughout other forums
Finished the cabinet. Ordered an unfinished cabinet from TRM and finished it myself. Dovetail jointed solid 3/4" pine.
Marked and drilled holes for chassis and handle. Block sanded by hand. Applied Watco Danish Oil Dark Walnut, 2 coats with a day in between. Then another 2 days before applying 2 coats of satin polyurethane clear varnish. After this had dried a couple of days used brown 3M abrasive pad to final smoothness and applied furniture paste wax. Plan to cover panel on amp with grill cloth to match speaker.
Had to purchase hardware separately. Corners, feet and handle were installed.
Here it is on the 2x12 speaker purchased from TRM several months prior to starting this ODS and finished in the same manner. I didn't know what the amp width was going to be when the 2x12 was ordered, amp is 5/8" wider than speaker.
20231022_151739.jpg
20231022_151732.jpg
Ordered resistor set from Jelle over the weekend and have another set of burned in and matched 6L6's currently out for delivery.
You do not have the required permissions to view the files attached to this post.
Received the other tubes yesterday, went home after work and put them in. Set bias to 40mA @ 437vdc. My calculation from that is 17.5W and 58%. Am I correct? 6L6 tubes.
Plugged in a guitar ( Epiphone 335 Dot with SD pickups) and before I knew it a couple of hours had passed. I really like this amp! Cleans sparkle, almost a faint reverb effect to them. Dirty has a nice mellow distortion to downright dirty crunch with good compression. Was only able to play moderately loud due to time of day so didn't really check out feedback.
Did notice Middle control not as effective as Treble and Bass. I'm open to suggestions of where to begin troubleshooting this minor issue.
GAStan wrote: ↑Tue Oct 24, 2023 4:45 pm
Received the other tubes yesterday, went home after work and put them in. Set bias to 40mA @ 437vdc. My calculation from that is 17.5W and 58%. Am I correct? 6L6 tubes.
This could be the reason of your noise problem.
