Cinema Engineering 7080 EQ: different input transformers - different response?

[plumsolly]

For the last couple of years I have been working on getting a pair of Cinema 7080 eqs running. I have finally got them to a point where I ran some frequency response tests and one of the units I have is about 3db down from the other at 20Khz. After some sleuthing, I determined the input transformers are the culprits. I isolated them from the rest of the circuit and with 27k between the secondaries, one is indeed about 3db lower at 20khz than the other. The case of each transformer looks identical, and the part numbers are the same, but the leads are different gauges. I am wondering if Cinema switched manufacturers at some point and I have two different transformers. I don't have any serial numbers to guess how far apart these unit were made, but there are some other subtle differences in components.

If you look at the schematic attached below, you can see C2 is there to bump up the high end (most likely to make up for transformer deficiencies). The reason the value is not marked on the schematic is that each unit I have seen pictures of has a different value (Cinema seemed to use a different combination of paralelled molded mica caps here to tune each unit?) I wonder if they just used a higher value for C2 with the crappier transformers? I had used 2200pF caps here originally. I put a 5800 pF in parallel and I can get a pretty similar response between the two units. Its a bit of a bummer as I was going to use them as stereo pair - I still will. I don't want to replace the transformer or pay Doc Hoyer to rewind one of them. If anyone has any ideas, I'd love to hear them, otherwise I'll just solder in that 5800pF and be done.

Thanks,

Ben

 

[emrr]

Reverse the secondary orientation to ground and see if it's better. 

 

[plumsolly]

Reverse the secondary orientation to ground and see if it's better. 

Finally had a chance to do this. The color of the secondary leads is reversed between the two units, so I was encouraged. But when I tested it, i get the same roll-off.

Any other ideas?

Best,

Ben

 

[emrr]

Variations of that are my best initial suggestion.  Do you see anything at all better or worse?  Enough to suggest you could make the other unit 'worse' to match by reversing that one?  Get them as  close with orientation swaps and then bring on the C compensation. 

 

[plumsolly]

Get them as  close with orientation swaps and then bring on the C compensation. 

Well, initially, reversing the secondaries didn't do anything. But then I brought in an input pad (ended up with 2.5K/510r maybe?) and all of a sudden it matters. So, a long story short, I ended up reversing the secondaries on both units and with some slight adjustment of C2 (~1700pf on one vs. ~2400 on the other)I have them close enough for my 2 bus. See below.

Now I need another ~8db to compensate for my input pad. Anywhere anyone sees might be a good place to do that?

Thanks,

Ben

 

[emrr]

R1-2 is a voltage divider.  You might shift those values to get more gain at the front, though you'd be undoing some of what your external pad is doing.  The next most obvious is R20-21, though likely more problematic due to various factors.  There may or may not be a lot of play between R18-21 and C13-14. 

 

[plumsolly]

R1-2 is a voltage divider.  You might shift those values to get more gain at the front, though you'd be undoing some of what your external pad is doing.  The next most obvious is R20-21, though likely more problematic due to various factors.  There may or may not be a lot of play between R18-21 and C13-14.

Was wondering about R1-2. Changing those won't affect the bias as there is no current going into the grid? I calculate 21db of attenuation. Seems like it would be easy enough to take that down to 13db or so, though I am wary of messing around in that area for obvious reasons.

Best,

Ben

 

[emrr]

No problem at all changing that relationship.