Franks Pultec EQP1 Sweep

Hello - I just completed this build with the Gyraf SRPP board and Franks Filter board. I took a sweep using RMAA, with no boosts or cuts and this was the result. I am a bit of a novice so I am not entirely sure if this result is normal or an issue that I need to fix or if they are negligible,

I noticed there is quite a large boost below 100Hz and there is a 0.25-0.4 db difference in the left and right channel, Im not sure how the noise and THD etc fairs against the benchmark?

so I would appreciate any insight anyone would be able to provide.

Totally normal.  Hardly a big boost in tube/transformer/NFB world.  Gain difference is standard for tube or parts variance, you should be able to find a way to trim that closer. 

emrr said:

Totally normal.  Hardly a big boost in tube/transformer/NFB world.  Gain difference is standard for tube or parts variance, you should be able to find a way to trim that closer.

Click to expand...

Oops I was supposed to post this topic in "The Lab" I some how posted it here. Sorry.

Thanks so much for that insight Doug, much appreciated. I can breathe a sigh of relief. There aren't any trimmers on the boards to match levels with, so I guess I just have to make do.

After giving this some more thought and giving in to my OCD - I have been wondering if it would be better to try and trim these two channels closer together?

I also understand that it has been mentioned that with tube transformer gear that my sweep response is "Totally normal.  Hardly a big boost in tube/transformer/NFB world.  Gain difference is standard for tube or parts variance."

I am a bit of a n00b so please excuse me and my OCD - When I look at the response for Lundahl transformers (LL5402) they seem pretty flat  so I just wanted to make sure that this response curve is ok and that "hump" in the low end is not something wrong.

It would be great if someone would be able to guide me as to how I can go about adding a trimmer to get these channels at the same level - or if the difference is negligible and not worth the trouble?

This is the schematic of the SRPP board
http://www.gyraf.dk/gy_pd/pultec/gy_pd_sch.gif

RMAA is a good tool but it relies on your soundcard for input. I would not be surprised to see a 0.25dB difference in the channels of a sound card. One way to check this is to use a splitter cable to send an identical signal to both channels of the sound card and then run RMAA. If you want to be true to your OCD you need to be sure the test equipment is better than what you are measuring!!

Cheers

Ian

ruffrecords said:

RMAA is a good tool but it relies on your soundcard for input. I would not be surprised to see a 0.25dB difference in the channels of a sound card. One way to check this is to use a splitter cable to send an identical signal to both channels of the sound card and then run RMAA. If you want to be true to your OCD you need to be sure the test equipment is better than what you are measuring!!

Cheers

Ian

Click to expand...

Hi Ian, thanks for taking the time to reply - thats a good point. I will try and get a splitter and test it out. I am using a Prism Sound Orpheus, so I'd like to think its a very close match, but lets see - I will post the results! Otherwise,  do you think the frequency response is ok, with that little bump in the low - using Lundahl transformers LL5402 with matched Amperex Tubes

I don't see how a half-dB up at 50Hz is "bad" in any way.

Aside from any bump in the amp+transformer, there is a fair chance the sound card does a software "zero correction" which nulls the DC, is sure to cause a slope in deep bass, and very possibly a bump/drop. So it could be the sound card.

+0.5dB up at 50Hz is IMHO good for most work. We rarely have even 3dB accuracy from speakers and rooms in this area, and we rarely complain "too much 50Hz!" (unless it is UK hum). Up-at-50Hz would scare me in that it might lead to rising sub-sonic response, but your graph clearly shows a steep cut at 20Hz.

The only time I had to think about a similar curve I had pipe-organ in an urban stone church. The faraway truck rumble was audible and even strong at 20Hz. I scanned each track for the lowest-pitch pipe used, and set a steep filter there. Long passages with no heavy pedal I set an even higher cut-off. But that's an unusual case, and was perhaps not warranted.

What I see in your graph is a 2.5dB bump at about 20Hz. It is important to remember that the Pultec design is from the era where everything was true 600 ohm source and load. By testing it with your sound card you are driving it from a source impedance which is probably considerably less than 600 ohms. One of the effects of this is that the the resonance of the transformer inductance and the sound card output capacitor is more pronounced and I suspect this is what is causing your bump. I would suggest you try driving it with a true 600 ohm source (just add a 270 ohm resistor in series with the hot and cold leads) and repeat the test. The gain will be about 6dB less but I expect the bump will disappear.

Cheers

ian

there could also be some difference in your bass boost caps,

1.5 uf is probably rated at 10 percent,