Stanton DJF

[jonesy]

Hi All,

I have a Stanton DJF-1 that I'd like to tweak out a bit. Its a dual 4 pole filter with line and RIAA inputs. Runs at -10 unbalanced. What I hope to accomplish is basically reduce or eliminate the noise floor, and bring the output up to +4.

The RIAA input is based on an LM833 and comes straight out of National's application guide for that chip: http://www.national.com/an/AN/AN-346.pdf). The output of the RIAA amp and the line input both pass through the line input buffer (TL072) and on to the filter circuit.

From looking at the schematic, it looks like each filter has its own input buffer (again LM833) and output buffer (TL074). Is that correct? When you run the filters in parallel (notch mode) the output of the filter sections is summed using another TL072. When you run the filters in serial, the output of filter 2 acts a the final output buffer. But what's interesting is that the schematic seems to indicate that the output is DC coupled... Basically What I want to do is balance the final output with a step up transformer (1:4 ???). How do I calculate the output impedance of the circuit? What's an acceptable buffer output impedance to be able to adequately push such a tranny? Will the transformer's output impedance be too high?

Finally, would there be any noticeable benefit to replacing the buffer amps with higher spec'd parts? Of course I'm going to do the obligatory coupling cap upgrade/bypass.

Also, looking at the power supply, am I dumb or does this thing have two rails? There are markings for +-12V and +-VV... ??? Why would they do that when all the opamps in the circuit can be run off one rail? But anyway, 12v would be rather low for the opamps right?

Schematic attached. Thanks in advance!

Mark

 

[abbey road d enfer]

Hi All,

I have a Stanton DJF-1 that I'd like to tweak out a bit. Its a dual 4 pole filter with line and RIAA inputs. Runs at -10 unbalanced. What I hope to accomplish is basically reduce or eliminate the noise floor,

That seems very difficult to do; most of the noise comes from the RIAA preamp and the state-variable filters. Improving them would involve a major redesign. Actually they are pretty well designed in an economical way.

and bring the output up to +4.

The TL0's are not happy driving a low impedance. You would need to add a line driver with about 12dB gain. I think you'd better use an active balanced circuit, which will give you the desired gain and cost a lot less than a good quality xfmr.

Also, looking at the power supply, am I dumb or does this thing have two rails? There are markings for +-12V and +-VV... ??? Why would they do that when all the opamps in the circuit can be run off one rail? But anyway, 12v would be rather low for the opamps right?

Bipolar rails is common good practive in audio circuitry. It has many advantages, in particular in the choice of coupling caps.

 

[jonesy]

Thanks for your feedback. You say most of the noise comes from the RIAA, but there is noise even if the RIAA is switched out of the circuit in favor of the line input. Not interested in modifying the filters themselves, I like the character, I just want to get a higher quality signal into and out of the box.

I understand bipolar rails are preferred, but from the schemo, it seems there are two bipolar rails... One at +-12v, and a +-VV. I'm looking at the pcb right now, and it looks like the +-VV rails are supplying the power to certain parts of the circuit, and the +-12v rails are are feeding other parts. I'm confused. The schemo doesn't show power connections and the pcb has traces on both sides so its a bit hard to trace. Tell me I'm being dense...

Anyway, it looks like the output is capacitor coupled, so that answers that.

Okay, active balanced. So I'd essentially be replacing the output stage and redesigning it from scratch? Any recommendations on where I can get smart on this? Example circuits, etc? Or could I potentially replace the tl0 with something that's more comfy driving a low impedance?

 

[jonesy]

Gobs of info all around for balancing line drivers! Thanks for pointing me in the right direction. It has also occurred to me that should be able to build an balanced to unbalanced converter and stick that in front of the line input. But there's not quite as much info around on that.

Anyway, are there any pre-made PCBs for these applications? A search around here shows that at least there used to be...

Still confused about the power rails. I'm picking up a DMM this weekend, so hopefully that will provide clarity. If the rails are indeed 12v, would it be a good idea to bump it up to 15v with a 7815/7915 and a 15v AC wallwart? Are the PS caps sufficient to handle this or should I bump them up too?

 

[jonesy]

So... INA 2137 on the input. I'll use it to drop the gain on the input by 6dB. This will feed into the input buffer, correct? Or will it take the place of that buffer and feed directly into the filter circuit (don't imagine so, since the phono preamp also makes use of the input buffer)? Coupling caps in front of and/or behind the INA2137?

A pair of DRV134s (didn't see any dual packages) on the outputs to balance and step the gain up by 6dB. Is this gonna sit after the output buffer, or should I pull its signal straight from the output of the filter circuit? Coupling caps?

Sorry for the newbie questions, iterative process here...

edit: looks like the THAT 1206/1606 would handle this application a little better because it offers higher CMR if you use it with unbalanced gear. Right now the DJF lives on the balanced inserts of a mic pre/eq, but I may also use it directly with my sampler, which has unbalanced outs...

edit2: okay so on the 1206, the cap that connects CMin to CMout essentially eliminates DC, rendering coupling caps unnecessary? And on the 1606, the two 10uF caps sitting between the out+/sense+ and out-/sense- (respectively) eliminate DC? So I can run from the input to the 1206 and into the TL072 line in buffer, and from the line out buffer to the 1606 and forward, without needing caps in between? Am I reading this wrong?