Output Transformer issue.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Mylithra

Well-known member
Joined
Feb 12, 2008
Messages
184
Location
So. Texas
I have a Vox AC15CC1 that I took in trade with the intention of gutting it and doing a Turret board circuit inside.  The original one had an issue where it would make odd distortion on the note decay coupled with some pops. I thought the pops sounded like a bad resistor and thought the odd distortion was something in the inter-stage coupling and since I was going to rip the old PCB out, I didnt take diagnosing any further.

Fast forward a couple of weeks and the new circuit is installed, an additional tube for the Top Boost EQ, and its sounding great. I didnt hear any of the popping or odd notes on the decay.

I lent the amp to a friend who is notorious for finding bugs in work through playing and he found a bug.  When he would do this dissonant  bend around the 10-12 fret with the high E and B, it would make this crackling + distorted sound over the notes. The only component of the original amp in this thing are the transformers. I replaced the output transformer and its no longer making the noise it was.

I'm not sure how I could have found this earlier in my troubleshooting process. If the transformer has outright failed, I can find that, but when its just sort-of failed as this one was, Im not sure how I could have found it earlier. Also, what kind of failure in the transformer could cause this kind of noise. I do have a signal gen and a o-scope if that would make the process easier but I don't know what to look for beyond the generated wave being reproduced faithfully.
 
I would expect intermod over transformer issues, interesting. 

If the winding is about to go, it can make crackling noises but those tend to be constant. 

I've seen output transformers that made popcorn noises at expected current draw, which disappeared with current limiting or if bypassed with an alternate B+ source so as to remove current flow. 
 
Mylithra said:
When he would do this dissonant  bend around the 10-12 fret with the high E and B, it would make this crackling + distorted sound over the notes.
Two close notes produce intermodulation products of low frequency. If the inductance of the transformer is low, the tubes fart out. The transformer may not be really faulty, but simply undersized.
Also unbalanced current can cause a reduction of inductance.
 
If I was laying the 2 transformers side by side, the replacement has a smaller lamination stack than the old one but the other dimensions to the transformer were the same. I did have to drill new holes about 3/8 of an inch closer to make 2 of the bolts fit. This amp when it was owned by my friend didn't do this until relatively recently (I had played with him for about a year before this amp started with the original behavior I described in my first post) so something changed between then and when it started doing this.

Id probably have to tear it down to see what exactly is wrong with it but beyond what Ive seen CJ do here, Im not sure I'd know what to look for.

 
check for mechanical first by banging on the side of the amp with your hand,

sometimes resonant frequencies of cabinets will jangle loose joints, but those notes seem high.

i just built an AC 15 , well, the part with the EF86, no vibrato or 12AX7a section,

used a 100 EI stack for the transformer, which is 1" x 1" core and have no problems,

gutted a PV bandit  one 12" for the chassis,

used some EL82 tubes instead of EL84 since Bugle Boy prices were crazy, got a pair for 20 bucks, Ultron brand, which distributes European brands,  so who knows what they are.

have you done a cap job? filter caps can sometimes cause weirdness,

using the choke pi filter in the supply? 6CA4 rect?

using non lytic caps in the pwr supply, sorry for the hijack  ??? >

 

Attachments

  • ac1.JPG
    ac1.JPG
    317.5 KB · Views: 28
EF86 problems are no joke, they happen all the time and those tubes ain't cheap, Amperex Bugle Boy being used, so we drop thr B+ a bit and add ohms to the screen dropping resistor,

1 meg to 1.5 meg, drops screen from 70 from 60 volts,

we took a chunk of sheet meta and cut a 3/4"  hole in it,  cut it down to size and screwed it to a piece of 1/4" neoprene, which then gets screwed to the chassis for a shock mount, using loose wires to connect instead of components,
 

Attachments

  • ac2.JPG
    ac2.JPG
    211.2 KB · Views: 24
Mylithra said:
If I was laying the 2 transformers side by side, the replacement has a smaller lamination stack than the old one but the other dimensions to the transformer were the same. I did have to drill new holes about 3/8 of an inch closer to make 2 of the bolts fit. This amp when it was owned by my friend didn't do this until relatively recently (I had played with him for about a year before this amp started with the original behavior I described in my first post) so something changed between then and when it started doing this.

Id probably have to tear it down to see what exactly is wrong with it but beyond what Ive seen CJ do here, Im not sure I'd know what to look for.
Are you sure the output tubes are matched? Do you have current probes? If not you can connect a 10 ohms 5W resistor in series with the cathode. Check that the voltage from the common resistor to both cathodes is reasonably identical. Typically 0.4Vdc +/- 10%.
You may have to adjust the value of the common cathode resistor in order to set the bias current to about 40mA.
 

Attachments

  • current probe AC15.jpg
    current probe AC15.jpg
    126.3 KB · Views: 11
mjrippe said:
Short article that may be relevant: https://www.mercurymagnetics.com/dealing-with-ghost-notes-courtesy-of-doug-roccaforte/
I do hate it when supposedly technical companies can not get the simple things right.....
"For 100 watt Marshalls expect to add another 100-150mf across the main B+, "
 
Walrus said:
I do hate it when supposedly technical companies can not get the simple things right.....
"For 100 watt Marshalls expect to add another 100-150mf across the main B+, "
Not only that:
"tubular capacitors are less prone. This is because they do not have any flat sides for the signal to bounce off of."
Certainly one of the most ludicrous assertions I ever saw...
 
abbey road d enfer said:
Not only that:
"tubular capacitors are less prone. This is because they do not have any flat sides for the signal to bounce off of."
Certainly one of the most ludicrous assertions I ever saw...

Yeah, I was not endorsing the bullshit part of that article, just the underlying theory.
 
abbey road d enfer said:
Actually, there's no theory to speak of in this article. For me it's just old wives tales.

So increasing filter capacitance will not reduce ghost notes?  You are probably right, just replacing the original caps that have degraded probably makes more difference.
 
mjrippe said:
So increasing filter capacitance will not reduce ghost notes?  You are probably right, just replacing the original caps that have degraded probably makes more difference.
Ghost notes is just one case of farting out, due to increased ripple of B+, which results in 120Hz (100 in Europe) superposed to the output signal.
Ther are many other causes that need a more analytical approach.
 
EL82's are running at about 260 V on the plates , maybe 256 on the screens, 

we have a Fisher 100 OPT in there which runs at about 6.5K : 8 Ohms, this thing breaks up at around 5 on the vol knob, so we use it in conjunction with a Princeton Reverb to get a blend of clean and dirty. It sure sounds nice for slide guitar  when it gets into the crunch zone.

there is a series resistor after the 6CA4  rectifier in the AC 10, probably to limit power to the smaller speaker, so we are using about 75 ohms there. this helps save the rectifier tube which is probably not as strong as a 5Y3 or 5U4.

doing a bend on one string while holding another string at constant pitch will certainly generate ghost notes if that is what your doing.  depending on the guitar, pickups, etc. 

notice that the C 15 uses a pi filter before the B+  so that choke could be doing weird things also. you rarely see this, i think i have only had one bad choke so far.
 
In the Vox AC15CC1, there wasn't an EF86. It was just 2x 12AX7s and 2x EL84s. There was no tube rec either. I know in the original AC15, there was.  In the version I put in this thing, I also didn't use a tube rec. Just a 4x 1N5399s in a full wave rec. (not bridge). Since I didnt have the capacitance limitations of the tube rec, I'm using a CE Mfg. can cap 40x40x20x20. I'm using 40uf for the primary B+ and another 40uf for the screen supplies. If it was a capacitance issue it would have been cured by the refit since I replaced all caps with new and there's more of it than there was before the refit.

As I mentioned before, the amp did a similar noise before the refit as well as after.
The only parts I kept of the old amp were the transformers and tube sockets.
Replacing the transformer did fix the issue.
I'm just trying to figure out specifically why.

 

Attachments

  • 20200103_130113_2.jpg
    20200103_130113_2.jpg
    1,016.9 KB · Views: 15
EmRR said:
If the winding is about to go, it can make crackling noises but those tend to be constant. 

I've seen output transformers that made popcorn noises at expected current draw, which disappeared with current limiting or if bypassed with an alternate B+ source so as to remove current flow.


Arcing on the list. 
 

Latest posts

Back
Top