|
Post by wiz on Aug 5, 2017 16:23:54 GMT -6
Hey Brad
great work and thanks so much for sharing.
Running frequency sweeps like this is an extremely valuable way to learn your gear.
The first time I did this on a pultec (warm) I went.. .ahhhhhh so thats why pultecs do what they do.
I run these types of tests on all my outboard to see what is what.. . running a test tone through them as well and watching it on an analyser, whilst playing with the level of the test tone and seeing the harmonics is great as well.
On my delta I ended up putting in a variable HPF in place of the static one ( i swapped out Aux 4 ) and using it and the two sweepable frequencies, I can get into pultec territory, enough for me that I actually just sold my Warm one....as thats how I mainly used it, to sculpt in that way you show above.. I found the high end boost scratchy, for want of a better term ( I also heard that in the Neve 551 I had and sold for that reason).. how are you finding the sound of the high end boost of the KT?
cheers
Wiz
|
|
|
Post by kcatthedog on Aug 5, 2017 16:48:02 GMT -6
Brad , I was wondering if the type of metal could/ does affect the resistance and inductance values ?
I remember there was a fair bit of talk about this when the Warm eq came out ?
|
|
|
Post by BradM on Aug 5, 2017 17:00:42 GMT -6
Yeah those certain high boosts don't look right. Looks like maybe they cheated and used fewer inductor taps to land all the frequencies. Series resistance is part of it, but that's not what we are seeing IMO. You can land any close frequency with different caps but the Q changes. I am on a phone, will look closer later. See my links to other measurements. Awesome! Thanks for your help. So are you implying that the reactance is off? I'm curious to hear more about why you think the Q is off? It's hard to tell how many pins are on the inductor without taking the unit apart. I'll check out your link. Founds some plots of the Cartec EQP1A curves: www.soundonsound.com/reviews/cartec-eqp1aBrad
|
|
|
Post by BradM on Aug 5, 2017 17:02:12 GMT -6
Brad , I was wondering if the type of metal could/ does affect the resistance and inductance values ? I remember there was a fair bit of talk about this when the Warm eq came out ? I really don't have much expertise as far as magnetics go. Maybe others can give more insight? Brad
|
|
|
Post by kcatthedog on Aug 5, 2017 17:15:42 GMT -6
its ok, was just curious, about what is sonically affecting what , for the guys and gals who really know their pultecs this was apparently a big deal.
|
|
|
Post by drbill on Aug 5, 2017 17:33:40 GMT -6
Sometimes it's awesome to be blissfully ignorant, and ignore the internet chatter. Sounds great to my ears, whether the Q is off or not.
|
|
|
Post by Guitar on Aug 5, 2017 17:51:08 GMT -6
Could be intentional? Who knows..
|
|
|
Post by EmRR on Aug 5, 2017 20:54:07 GMT -6
It's not intentional if you're copying a Pultec! There are old books full of charts on arriving at correct L and C values for particular Q's. You can arrive at a particular frequency with a lot of different combos, and they will all give different Q's. You can't do anything with resistance other than lessen or extend the range if I recall, it doesn't affect shape directly. Caps are cheap and easy, I'm sure it's caps that are being changed as expected, but they might be using fewer inductance taps which would skew the Q's at some points.
At one point I designed out the 'super-Pultec' that did every damn combo available and a bunch of things they don't currently have controls for, and it's too many knobs! (16, and three 3 position switches) Would need a laminated guide hanging from a string off the front.
|
|
|
Post by EmRR on Aug 5, 2017 21:02:33 GMT -6
The 16k, 12k, and 10k Broad boosts seem to have a much wider Q than the 8k, 5k, 4k, and 3k boosts. That 10k setting (orange) is super broad compared to 8k (green). This doesn't look right to me, especially when I compare to the plots published in the vintage Pultec datasheet/manual. Does anyone have access to plots of the HF boosts for other Pultec EQ's on the market? It would be interesting to compare. 3/4/5K in particular look closer to the dog whistle Q's found in the mid frequency Pultec than the program Pultec. A friend brought a Pultec he'd built from scratch with a mix of inductors from various places rather than a custom multi-tap meant for a Pultec, and the Q's of the various bands had no consistency at all, it was all over the place from one frequency to the next. Here's all mine again: groupdiy.com/index.php?topic=48328.msg703839#msg703839and in particular: Full size plot herefarm4.staticflickr.com/3682/12319170175_33e29bec0a_b.jpg
|
|
|
Post by BradM on Aug 5, 2017 21:10:02 GMT -6
It's not intentional if you're copying a Pultec! There are old books full of charts on arriving at correct L and C values for particular Q's. You can arrive at a particular frequency with a lot of different combos, and they will all give different Q's. You can't do anything with resistance other than lessen or extend the range if I recall, it doesn't affect shape directly. Caps are cheap and easy, I'm sure it's caps that are being changed as expected, but they might be using fewer inductance taps which would skew the Q's at some points. At one point I designed out the 'super-Pultec' that did every damn combo available and a bunch of things they don't currently have controls for, and it's too many knobs! Would need a laminated guide hanging from a string off the front. I bet you are right - they went with less taps on the inductor and then tweaked the cap values. I'm going to have to take the thing apart and find out. Yeah, compared to your plots the Q for 3-10k are wrong. Booo! Brad
|
|
|
Post by EmRR on Aug 5, 2017 21:14:19 GMT -6
And again, my plots are an Analog Allstars unit, but the curves seem to match both what's in the original Pultec documentation AND the plots Manley publish, so I think they got it extremely close to right. You can see the little bit of Q difference every other frequency, and that's the result of inductor tap sharing. Original Pultecs did a little of it, and clones with additional frequencies do more of it. I don't know of a Pultec clone with expanded frequencies which also expands the taps on the inductor with a custom unit; there may be one.
I expanded the frequency selection on the ioaudio mid Pultec, only option being additional caps, and the additional curves change Q dramatically.
|
|
|
Post by BradM on Aug 5, 2017 21:18:14 GMT -6
And again, my plots are an Analog Allstars unit, but the curves seem to match both what's in the original Pultec documentation AND the plots Manley publish, so I think they got it extremely close to right. You can see the little bit of Q difference every other frequency, and that's the result of inductor tap sharing. Original Pultecs did a little of it, and clones with additional frequencies do more of it. I don't know of a Pultec clone with expanded frequencies which also expands the taps on the inductor with a custom unit; there may be one. Doug, Who makes the inductor for the Analog Allstars and Manley units? Cinemag? Brad
|
|
|
Post by EmRR on Aug 5, 2017 21:21:24 GMT -6
I do not know, I doubt Cinemag. I think Manley wind their own, and Analog Allstars may also be hand winding theirs, which is a correct old school toroidal style.
|
|
|
Post by BradM on Aug 5, 2017 21:29:54 GMT -6
I'm a little disappointed KT didn't spend the extra $1 to get this right. There are 5 pins on the inductor in the unit. There should be 7. So that implies that two of the frequencies have an unoriginal inductor/cap combination. Hmm...So why then are most of the HF curves off then? Did they just botch the entire inductor? I guess I'll have to disassemble and measure the inductances and see what caps are used... Yay - a project.
Brad
|
|
|
Post by EmRR on Aug 5, 2017 21:49:51 GMT -6
Hmmm. They blew something if accurate clone was the goal. Analog Allstars is 7 tap. You can't get most people's idea of a Pultec right with 5. The early EQP-1R version only had 3 taps though. There was a version with 6 if one drawing is correct. If you have expanded high frequency options you need 7 taps as the starting point and a proper expansion design would probably go to 8 or 9 taps.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 6, 2017 3:20:10 GMT -6
When KT put this thing out they weren't reckoning on guys like Brad and EmRR looking under the hood and putting it through it's paces .... people like it for $300 - would they feel the same for $6-700 ?
|
|
|
Post by jcoutu1 on Aug 6, 2017 7:54:29 GMT -6
I'd love to see similar tests done to the Warm. I bet the high Q is done similarly.
|
|
|
Post by EmRR on Aug 6, 2017 8:19:01 GMT -6
A look at the Warm would be interesting, has it not been done by someone somewhere? I'd think they are using the stock Cinemag inductor, which would be correct. Pictures show Cinemag parts.
|
|
|
Post by jacobamerritt on Aug 6, 2017 9:22:27 GMT -6
Pictures show Cinemag parts. Don't the Warms just use Cinemag transformers? I also would love to know if the Warms are 5 or 7 tapped inductors.
|
|
|
Post by Ward on Aug 6, 2017 9:48:27 GMT -6
Okay now we are getting to the good stuff. This is where the Pultec really reveals itself as a very complex EQ capable of some very versatile EQ curves depending on where you have the knobs set. If you are courageous enough to use your ears and turn the knobs, you will be sonically rewarded. I find using a hardware Pultec is superior to plugins in this instance because I can turn both boost and cut knobs at the same time with two hands. That makes it much easier to get where you want to go. For this plot I took measurements of 20, 30, 60, and 100 Hz with both the LF Boost and Cut turned up. For all curves the Boost was set to 10. Cut is shown set at both 5 and 10. You can see that as you vary the amount of Cut the frequency of the dip actually shifts. The numbers "20", "30", etc. are really meaningless when you are using a Pultec like this to do the "low frequency trick". Note, that I forgot to capture the curve at 100 Hz with Cut at 5. Again, the frequency markings mean nothing in these cases. If you get the ratio of LF Boost and Cut just right you can actually use those controls to give you what are effectively midrange dips: Brad Thank you so much for posting all this fantastic info!! The "mid dip" from the low boost and cut is precisely what many of us refer to as the "low end alignment" that a Pultec does. The warm does this too. Glad to see the KT can also pull it off.
|
|
|
Post by EmRR on Aug 6, 2017 11:40:52 GMT -6
I never did understand how they came up with the frequencies they marked on the faceplate for this design. The numbers don't really correspond to any -3 dB or midway point on the curves. -3 dB wasn't any sort of standard back then. The labeled boost frequencies seem to relate to the point that is -1dB from the boosted shelf plateau, which is along the curve on the way down the boost rise. The exact frequencies are always different from the nicely fudged numbers printed on the panel too.
|
|
|
Post by ChaseUTB on Aug 6, 2017 13:47:29 GMT -6
I'd love to see similar tests done to the Warm. I bet the high Q is done similarly. How much do you want to bet 😫😀I believe warm is more accurate than KT .. I would love to be wrong considering price point ( not that a stereo pair of pultecs is my next buy but still ) call me a fool but I do believe warm does go the extra mile.. like revising their units to include the gain trim pots when they did not have to..
|
|
|
Post by BradM on Aug 6, 2017 13:56:56 GMT -6
How much do you want to bet 😫😀I believe warm is more accurate than KT .. I would love to be wrong considering price point ( not that a stereo pair of pultecs is my next buy but still ) call me a fool but I do believe warm does go the extra mile.. like revising their units to include the gain trim pots when they did not have to.. Measurements on the Warm coming shortly. Stand by... Brad
|
|
|
Post by EmRR on Aug 6, 2017 13:57:56 GMT -6
I'm a little disappointed KT didn't spend the extra $1 to get this right. There are 5 pins on the inductor in the unit. There should be 7. So that implies that two of the frequencies have an unoriginal inductor/cap combination. Hmm...So why then are most of the HF curves off then? Did they just botch the entire inductor? I guess I'll have to disassemble and measure the inductances and see what caps are used... Yay - a project. Brad Brad, have a look at the relative shapes at lower levels, like +6dB, etc. See if there's a level at which they seem indistinguishable, and if there's a level at which they begin to visually/audibly deviate from one another.
|
|
|
Post by Guitar on Aug 6, 2017 14:11:51 GMT -6
How much do you want to bet 😫😀I believe warm is more accurate than KT .. I would love to be wrong considering price point ( not that a stereo pair of pultecs is my next buy but still ) call me a fool but I do believe warm does go the extra mile.. like revising their units to include the gain trim pots when they did not have to.. Measurements on the Warm coming shortly. Stand by... Brad Woohoo! You're really on it! Somebody should be paying you to review and test gear.
|
|