[nikongod]

 
Quote:

Do you have some more impressions of this amp? It would be appreciated.

I swapped the transformers with a better quality set I had laying around and secured them in the most make-shift fashion imaginable (zipties). I was initially impressed by the SEX, but like many upgrades some things only make sense looking backwards. I am MUCH happier with the amp with these upgraded transformers. The overall detail of the amp went up a few steps, Very nice!
 
The new transformers are 10K:300,32 transformers with a dedicated primary/secondary, wired as autoformers like the stock amp. I am using only the 32ohm tap. With both these and the original transformers this amp has a fair bit too much gain for my tastes which Im not really a fan of. I have not measured the gain of the amp (or much of anything on the amp for that matter. I did measure the transformers by themselves and they are pretty sweet.) but I would guess that the gain of the amp is on the high side of 10.
 
I'l post some pics of the amp with the new transformers soon.

 

[MrTom]

Fantastic, loking forward to more photos!

 

[nikongod]

I found a moment to measure the amplifier the other day, and took photos at a mini meet at Larry's house. 
 
The measurements with the new transformrers are not outstanding, but formidable. 
 
The unloaded voltage gain is 10 when the 32ohm tap of the transformer is selected, with the 300ohm tap gain is 31.
 
The amplifier will swing about 20vp-p on the 32ohm tap and ~62vp-p (electrostatic much?) non the 300ohm tap. 
 
Output impedance on the 32ohm tap is 14ohms, BEFORE the addition of the 120 ohm output resistors on the headphone output.
 
Im not sure why I wrote it down but the amp swings about 405vp-p as measured at the top of the autoformer. 
 
The -1db bandwidth of the amp is not particularly impressive at 60-15.5Khz (10vp-p into 32ohms, across the speaker taps - no 120 ohm output resistor) I suspect the low end rolloff is due to a coupling cap somewhere, and the high end is a combination of the transformer and miller capacatnces around the amp. The bandwidth did not change significantly at lower levels.
 

 
The NHT PVC is to skim off a few db before the amp. My home source has 4v outputs and gain of 10 is unbearable. More gain is more important than 2 attenuator in series.
 
My plan at the moment is to disconnect the gain triode and let the power triode do its thing by its self. Im on the lookout for tubes that may fit the circuit though

 

[Fitz]

Good save, Ari.
 
With SEX it's always a bad thing when something breaks deep inside. Glad it didn't end in even more trouble.

 

[Zigis]

Quote:

The NHT PVC is to skim off a few db before the amp. My home source has 4v outputs and gain of 10 is unbearable. More gain is more important than 2 attenuator in series.
 
 

Simple way is soldering between input sockets and pot resistors, 3-4 times larger than pot's resistance.

 

[Beefy]

From memory, doesn't the amp run an 8ohm tap by default?
 
Quote:

Originally Posted by nikongod /img/forum/go_quote.gif
 
The unloaded voltage gain is 10 when the 32ohm tap of the transformer is selected, with the 300ohm tap gain is 31.
 
The amplifier will swing about 20vp-p on the 32ohm tap and ~62vp-p (electrostatic much?) non the 300ohm tap. 
 
Output impedance on the 32ohm tap is 14ohms, BEFORE the addition of the 120 ohm output resistors on the headphone output.

 

[nikongod]

 
Quote:

Simple way is soldering between input sockets and pot resistors, 3-4 times larger than pot's resistance.

I like shunted pots, and other things like this, but have other plans for the amp before I get there.
 
The NHP TVC is a temporary solution in the middle of what feels like its going to be a longggg build

 
Something that I did notice with the 100Kohm pot agrees with something someone mentioned to me on the subject of pot matching: I was chatting with a DIYfriend about 6 months ago who matches pots obsessively for small amps he likes to build, and he commented that the worst error was always in the first 100-150ohms of resistance. I asked him if higher impedance pots had better matching than low impedance pots because of this, and he confirmed that they do. The pot in the amp appears to be a fairly standard 100Kohm 24mm carbon pot (you know... your standard ~$3 audio taper pot) but has very good channel matching. Its just something I never had a method to verify for myself so I may as well write it down somewhere. 
 
Quote:

From memory, doesn't the amp run an 8ohm tap by default?
 

 
Yes it does. 

 

[nikongod]

I didnt really like having the transformers exposed like they were, but was anxious to hear them so made something work temporarily. There is no DC on them, but the AC voltages can be *cough* significant. 
 
I decided to get little covers for the transformers and try to DIY pot them. There was a lengthy thread on this topic some time ago, If I recall about when Voodochile built his MAD clone amp. I think the ultimate conclusion was that even a mildly competent job, with bubbles and all would perform somewhat like a properly potted transformer.
 
I found that devcon 2-ton epoxy lists "potting electronics components" as one of its applications in its datasheet, and also lists electrical dielectric strength. This might not be *exactly* what they had in mind, but the transformers dont heat up and the voltages are WAY below anything like what the datasheet lists.
 

 
How can you resist the cuteness? Seriously! 
 

 
The little boxes I found are a perfect fit for the transformers and a full 25cc container of epoxy. maybe 30-35cc would be better but I didnt want to cover any of the connections on the transformer. 
 
The notches in the sides of the boxes are to allow the boxes to sit over the wiring grommets in the top panel of the amp.
 
I only put these together an hour or 2 ago so I'm going to let them cure overnight and put everything back together tomorrow. 

 

[n_maher]

Outstanding work Ari, and a heck of a fun read to boot.

 

[nikongod]

I was anxious to get the amp working again today so I got cracking.
 
Reinstalling the transformers was an uneventful task aside from the fact that there is very little room to work in the amp and I broke the 2 non-broken corners of the base. The nuts holding the transformers and plate chokes are located in a space that is just a few mm narrower than my fingers. It would also help if I had a third or fourth hand to hold the stack of washers, screws and nuts that were necessary for putting everything together.
 

 
I upgraded to white tape! The new transformers revealed the subtle, but obvious, coloration of the blue tape. You are all a bunch of deaf fools if you cant hear the superiority of white tape. It makes sense, white is all of the colors of the spectrum, not just blue. 
Seriously: while I was working on the amp the wood base got knocked off of the shelf it was on and the 2 corners that were previously OK became "not OK" and the blue tape broke. Thankfully the actual amp was not in the base of this would have been a disaster of epic proportions. I had originally planned some mods that would require drilling into the wood to mount big heavy bits of iron N_Maher Menace_TC_lite style but I have since realized that the power transformer on this amp is already running pretty much topped out, and the mods I had in mind would take more current. Another project to add to the list

 

 
Aww so cute! You can see the little notches I filed on the bottom there. 
 
While I was installing the transformers I added 36ohm resistors across the secondary of the transformer to try to load the transformer better, and suck off a bit of gain. It sounds a touch nicer, but still has way a lot of gain.