Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › A Very Compact Hybrid Amp
New Posts  All Forums:Forum Nav:

A Very Compact Hybrid Amp - Page 84

post #1246 of 2201
5/10/09 at 4:40pm
1. Is your amp quiet or noisy

12au7 near silent especially leaving it on for a while
6n19 annoying hum

2. What enclosure are you using

longer version of the default case

3. What coupling caps are you using

Russian Teflon FT3, air wired. The wire is covered with copper tubing, this cut majority of the hum with 12au7 as the exposed wire was picking up signals.

4. Where have you located the off board components and their wiring

input / output run from under the board to the front. Planning to cover them in copper tubing soon.

5. Is your pot grounded

no, hum is still present with or without.

6. Have you done anything else unique to your build

used shielded tube adapters, used stepped pot, c4 caps is outside of the case, the reg is attached to the case for better cooling solution,

7. What headphones

hd600
yh-1
k501
koss pro/4aa
Reply
post #1247 of 2201
5/10/09 at 6:22pm
Quote:
Originally Posted by wiatrob View Post
However, should there be AC on the OUTPUT of the switching reg (measured from SG)? I'm measuring ~12 and 14VAC (6.3/12.6V Heater) on pin 1 of the 2595...
There will be some, but after the post-reg filter, it should be very small. That is your ripple voltage (AC remnants), and it should be at 2x your AC Hz. Rectified AC is @ 120Hz in 60Hz countries (like US) or 100Hz in other places with 50Hz AC. Your DMM will likely be in the error range at that point. You'll need a scope to get an accurate reading and if it's in the uV range it would be best to amplify it (with something like Tangent's LNMP).

However, if you're getting 40mV (like what adamus reported) it's well within the range of a good DMM and will show as that's a gigantic post reg filter ripple.
Reply
post #1248 of 2201
5/10/09 at 6:35pm
Hmm, I'm buying the DMM error on the output of the reg. Measured after L1H, less than a mV (with Fluke 189). I'll try to bring a scope home tomorrow.
Reply
post #1249 of 2201
5/10/09 at 6:51pm
Thread Starter 
Quote:
Originally Posted by holland View Post
There will be some, but after the post-reg filter, it should be very small. That is your ripple voltage (AC remnants), and it should be at 2x your AC Hz. Rectified AC is @ 120Hz in 60Hz countries (like US) or 100Hz in other places with 50Hz AC. Your DMM will likely be in the error range at that point. You'll need a scope to get an accurate reading and if it's in the uV range it would be best to amplify it (with something like Tangent's LNMP).

However, if you're getting 40mV (like what adamus reported) it's well within the range of a good DMM and will show as that's a gigantic post reg filter ripple.
Maybe. If this were a power rail 40mV would be huge ripple and would buzz everything.

But this is a heater line. Amps are dead quiet with 12.6/6.3VAC on the heaters where the ripple voltage peaks are 9V and 17V.

I think that 100Hz/40mV at pin 4 of the tube will not find its way into anything.

Now, if this buzz is making its way to the ground plane or somewhere else then it is definitely a problem. If it is, we haven't found out how yet.
Reply
post #1250 of 2201
5/10/09 at 7:14pm
Quote:
Originally Posted by wiatrob View Post
Hmm, I'm buying the DMM error on the output of the reg. Measured after L1H, less than a mV (with Fluke 189). I'll try to bring a scope home tomorrow.
1mV, wouldn't worry about it...

Quote:
Originally Posted by runeight View Post
Maybe. If this were a power rail 40mV would be huge ripple and would buzz everything.

But this is a heater line. Amps are dead quiet with 12.6/6.3VAC on the heaters where the ripple voltage peaks are 9V and 17V.

I think that 100Hz/40mV at pin 4 of the tube will not find its way into anything.

Now, if this buzz is making its way to the ground plane or somewhere else then it is definitely a problem. If it is, we haven't found out how yet.
You're right, of course, that it probably doesn't matter on the heater line. I've got very limited experience with tubes, and just finished my Bijou (which is full on AC as you're aware), and it's dead silent with my D2000.

I have my doubts, as well, that it's ground contamination.
Reply
post #1251 of 2201
5/10/09 at 7:53pm
Hmm.

Don't mess with anything that is working.

 Well, I did something to make the hum worse, so if I can figure that out, maybe it's a clue. Paralleled two 100u's to C1,2H - no change. Upped L1H to 6,6, louder. Restored caps and L1H to original value, still loud, but L1h is tacked under the board.

EDIT: Found out where this louder hum is coming from. ICP dead again - outputting 27V. This happened to me early on in the proto phase with the exact same result. I'll replace it tomorrow...
Reply
post #1252 of 2201
5/11/09 at 9:06am
Quick followup. Back to where I started. 12V tubes are silent, very slight noise on 6V tubes (slightly louder with the 6H30). @4-5mVAC at pin 4.

Might everyone with a very noticeable hum measure output of ICP and make sure it's in spec?
Reply
post #1253 of 2201
5/11/09 at 10:24am
Just an FYI: My patient is on the operating table w/DMM + scope attached. Plan to get into this tonight.

BTW, wife's B-day was last Friday, had lots Saturday & Sunday was Mothers Day. Not able to get into this, let wife know about the flurry of posts happening about "my" part of the CTH design.... She asked "Don't these guys have mothers?". A good question, I had to admit, for which I had no answer
Reply
post #1254 of 2201
5/11/09 at 10:33am
Dude, we've got Moms! Just no wives with kids

My build is back to working - it's got 1000u on C3H and 200u at C1/2H (and it somehow made it back into the case)- without that much of a change in ripple at pin4 of the tube. Methinks the heater design was right on.

But what do I know! I'll grab a scope at work today time permitting...
Reply
post #1255 of 2201
5/11/09 at 10:51am
Quote:
Originally Posted by wiatrob View Post
Dude, we've got Moms! Just no wives with kids
He he. You should know my sense of humor by now A bit caustic. Did think it was funny for her to ask that tho

WRT the heater design - Like I do my real job, software development, I assume bugs are mine until proven otherwise. In this case, I'll only be convinced the heater circuit plays no role in this issue after my disabled switcher/battery heater test tonight.
Wouldn't be surprised find the heater circuit is involved, either through its EMI or its reception of EMI. Again, things are exceedingly tight on the CTH PCB, esp. vs my P2P POC:

that had input stage a good 4 inches or so from PS. And more space between SM heater portion (bottom) and HV section (top).

Edit: BTW, my suspicion as to why this did not really come up during 1st 10-PCB proto run is that 12au7s were more commonly used, as maybe higher-Z/less efficient cans. Most were also enamored w/the amps sound enough to possibly overlook this.
Reply
post #1256 of 2201
5/11/09 at 11:07am
Alex, here's the data on my prototype.

1. Is your amp quiet or noisy: No, very quiet with a 6dj8 or 12BH7a
(there was some hum until I replaced the unshielded input wires with shielded)

2. What enclosure are you using: Hammond 1602 with a perforated top cover

3. What coupling caps are you using: Wima ( MKIV's ?)

4. Where have you located the off board components and their wiring
RCA's in, 1/4 inch out and Alps RK27 all mounted on the front panel. RCA's to RK27 to board using shielded cable. Wires out to the 1/4 jack are unshielded.

5. Is your pot grounded: No

6. Have you done anything else unique to your build: No

7. What headphones: AKG K601
Reply
post #1257 of 2201
5/11/09 at 12:20pm
Quote:
Originally Posted by cfcubed View Post
He he. You should know my sense of humor by now A bit caustic. Did think it was funny for her to ask that tho
Make sure she knows National Hybrid Headphone Amp Day is coming up!

Quote:
Originally Posted by cfcubed View Post
WRT the heater design - Like I do my real job, software development, I assume bugs are mine until proven otherwise. In this case, I'll only be convinced the heater circuit plays no role in this issue after my disabled switcher/battery heater test tonight.
That's a refreshing change from the attitude of some of the developers I work with I hadn't realized the POC build was quite so spaced out. That might explain the Delta between battery/switcher. Look forward to your results - and saving me from doing it. I lifted about 6 traces goofing around yesterday.
Reply
post #1258 of 2201
5/11/09 at 12:29pm
Thread Starter 
I'm not convinced yet that the heater supply is at fault. That is, I'm not sure that there is anything wrong with its design. It could be, as you guys have noted, simply proximity issues or it could be some interaction that was not caught during prototyping.
Reply
post #1259 of 2201
5/11/09 at 1:13pm
Quote:
Originally Posted by wiatrob View Post
Hmm.

Don't mess with anything that is working.


EDIT: Found out where this louder hum is coming from. ICP dead again - outputting 27V. This happened to me early on in the proto phase with the exact same result. I'll replace it tomorrow...
Checked mine and its dead on 24v.

I also checked the heater supply again on the scope, looks good enough, and like alex says, in the bijou we are using ac, so the odd millvolt shouldnt have an impact.
Reply
post #1260 of 2201
5/11/09 at 3:22pm
Quote:
Originally Posted by MrSlim View Post
5. Is your pot grounded: No
MrSlim, can you explain why some builds need the pot grounded, and some do not (as in NO hum when you touch the knob)?
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › A Very Compact Hybrid Amp