AMS Neve 1073 Sweeps
- Thread starter Aaronrash
- Start date
Help Support GroupDIY Audio Forum:
You need to use the latest REW beta. The beta allows you to use ASIO drivers (assuming your soundcard has them) which allows you to select sample rate and which I/O you want to use.
After installation, you'll need to setup the preferences, then click "measure" in the upper-left of the screen. BAM!
After installation, you'll need to setup the preferences, then click "measure" in the upper-left of the screen. BAM!
ding
Well-known member
give this a try...
http://johnr.hifizine.com/2013/02/room-eq-wizard-on-the-mac-an-input-workaround/
What interface do you have? If you are going from a balanced +4 dBu to a typical mic preamp balanced input, all you need is a 40db pad. Or just turn down the interface output by 40db. I think this is right.
http://johnr.hifizine.com/2013/02/room-eq-wizard-on-the-mac-an-input-workaround/
What interface do you have? If you are going from a balanced +4 dBu to a typical mic preamp balanced input, all you need is a 40db pad. Or just turn down the interface output by 40db. I think this is right.
Yes...
You need to feed the signal into the EZ1073, then connect the output of the 1072 back into your audio interface.
For me, I have a line input that is just a pad in front of the mic transformer. I thought about that last night, Aaron doesn't really mention which input he's using or how much gain is applied.
You need to feed the signal into the EZ1073, then connect the output of the 1072 back into your audio interface.
For me, I have a line input that is just a pad in front of the mic transformer. I thought about that last night, Aaron doesn't really mention which input he's using or how much gain is applied.
I have to add I am thinking of starting a new ez1073 and doing the wiring this time- keeping the
soldering to pcb as minimal as possible, moving those transformers.
Interesting your nv73 measures the same TheJames. Again, I wonder how the ez1073/500 would result from a sweep.
cheers
thanks alot
Hay
soldering to pcb as minimal as possible, moving those transformers.
Interesting your nv73 measures the same TheJames. Again, I wonder how the ez1073/500 would result from a sweep.
cheers
thanks alot
Hay
Well...
I wired up the input transformer output hard wired to the switch with shielded wire, zobel cap right on the transformer, input wired to edge connector with shielded wire.
I'm not able to duplicate the response.
This is making me wonder if there's some special sauce in the iron.
I need to go back and re-read what Aaron posted about his initial EZ1073 experience.
I wired up the input transformer output hard wired to the switch with shielded wire, zobel cap right on the transformer, input wired to edge connector with shielded wire.
I'm not able to duplicate the response.
This is making me wonder if there's some special sauce in the iron.
I need to go back and re-read what Aaron posted about his initial EZ1073 experience.
Aaronrash
Well-known member
If you're using the PCB mount transformer, don't even bother, it doesn't sound correct.
Neither transformer is PCB mount. But I suspect there's a difference in the output transformer. Terminated or not, zobel or not, the response is pretty much the same.
Can you get the same AMS response with the Carnhill transformers? If so, which Carnhill output are you using?
Can you get the same AMS response with the Carnhill transformers? If so, which Carnhill output are you using?
Aaronrash
Well-known member
I only have AMS transformers and I suspect they got pissed about this thread because they no longer sell them.
Anyways, if you're not seeing differences between terminated and unterminated something is definitely off somewhere... Regardless of what transformer you're using
Anyways, if you're not seeing differences between terminated and unterminated something is definitely off somewhere... Regardless of what transformer you're using
ding
Well-known member
I would have to agree with Aaron here. If you can't hear a difference between terminated and unterminated there must be something else going on.
The difference in the iron here is not that great. Even the pcb mount iron had a very similar sweep to Aaron's AMS sweep. Look at page 1 of this thread reply #19. You can clearly see that even the pcb mounted transformer looks similar in frequency response to Aaron's original sweep. I don't want to start a gearslutz type debate here but from what I have read those output transformers are made of the same material and turned the same way and AMS has never denied that. They just say "well we can't comment on that but ours is to spec" but Carnhill has said there is no difference. If there was a difference AMS would be jumping all over that to say so.
I believe Aaron when he said that he got his ez1073 to sound similar to his 1084 but it makes no sence to me that it was because of the tx wiring bypassing the pcb. There must be something else here going on. Once I finish mine I will listen and modify to my needs.
I do a lot of mixing and to my ears the difference in Aaron's samples are in the mid-low and low end. The original ez1073 Aaron built sounded more hyped in the low end where as now it sounds less so with a tiny bit of mid forwardness. This has the appearance that the high end is more defined but I can tell you once you get rid of some low-mid mud the high end always seems to shine thru. The sweeps bear witness to that. I am guessing here but maybe it has to do with impedance. Just a wild and unfounded guess. Ah well enough of my nonsence.
The difference in the iron here is not that great. Even the pcb mount iron had a very similar sweep to Aaron's AMS sweep. Look at page 1 of this thread reply #19. You can clearly see that even the pcb mounted transformer looks similar in frequency response to Aaron's original sweep. I don't want to start a gearslutz type debate here but from what I have read those output transformers are made of the same material and turned the same way and AMS has never denied that. They just say "well we can't comment on that but ours is to spec" but Carnhill has said there is no difference. If there was a difference AMS would be jumping all over that to say so.
I believe Aaron when he said that he got his ez1073 to sound similar to his 1084 but it makes no sence to me that it was because of the tx wiring bypassing the pcb. There must be something else here going on. Once I finish mine I will listen and modify to my needs.
I do a lot of mixing and to my ears the difference in Aaron's samples are in the mid-low and low end. The original ez1073 Aaron built sounded more hyped in the low end where as now it sounds less so with a tiny bit of mid forwardness. This has the appearance that the high end is more defined but I can tell you once you get rid of some low-mid mud the high end always seems to shine thru. The sweeps bear witness to that. I am guessing here but maybe it has to do with impedance. Just a wild and unfounded guess. Ah well enough of my nonsence.
Aaronrash
Well-known member
Nope nothing else going, at least with my units... I have spent months of time messing with the EZ1073. I've tested every single part of the circuit. I've changed every component several times. I've tried lots of different transformers. I've tried modified power supplies. I've tried super high end components. Tants, even tried electros at one point, NOS BC184C... And all at the same time while building my own etched units, as well as owning a AMS 1073 and AMS 1084.
I've studied these units to the point where I'm sick of them. I know all the techniques and wired schems used in the originals. The grounding scheme is very specific in the original schematic. If you pay close attention you will start finding the differences in how the EZ1073 is wired up as well as the 2 resistors that are off from original value, I am surprised nobody has spotted those yet...
From what I've learned so far, the biggest impact on the sound of a 10XX class A module is the output transformer with the DC flowing through the primary and the gap at the core that is 0.003. The tolerances I believe are +/-20% with the Carnhill output transformer and probably even worse back in the day. I believe this is the reason allot of people say no two 1073s ever sound alike, even the same issue number and build date... It's that massive chunky LO1166, go back to what Rupert said in Sound City... that pretty much sums it up.
Don't let someone say it's components. Every single channel on the SSL G Series I work on sounds exactly the same and there's thousands of components on each channel. Capacitors and resistors are made to such strict specs today that those aren't to blame when things don't sound right. Transformers, are another story.
If you don't wire that output transformer correctly you will start getting stuff that doesn't sound right. It wasn't until I bypassed as much as the PCB as I could I started hearing the EZ 1073 units like I've never heard them before. If even 1 little tiny trace on the board is too close to another trace that it doesn't like, problems start happening. I'm not gonna say the EZ1073 board is flawed but I will say from my own personal experience with the 2 boards that were sent to me, they sound better with what I did to them, and clearly measure better too.
Even still at the end of the day, my AMS 1073 has something magic to it that sounds even better than my AMS 1084
(witch retails for a nice $4,295.00) Yeah $,4,295 for a preamp/eq... Anyways the AMS1073, it has a harmonic sparkle thing in the high end that even my 1084 doesn't do. My home etch units don't do it and neither do both of my EZ1073s.
All the same components in everything. Witch again really leads me to believe that those output transformers are one of the hardest things in the world to get perfect and even harder since a EM***** spec sheet was never made for the 1166. Leads me to believe my AMS 1073 has a really special output transformer in it by luck, and every single unit measures the exact same for freq and dist. This just makes even more sense to me when people say "something else is going on" It's just the variances in the output transformer we are hearing. The input transformer is pretty neutral. And come on, when you're paying AMS prices for preamps, don't you think they should all have the magic sound?
Go look at Avedis MA5 and you will start understanding why he has a custom Jensen output transformer made to his exact specs that he designed. He could have just used that Carnhill PCB output transformer...
Anyways, to sum it up. Experiment and more importantly get on to making music. This Neve quest thing has put a huge dent in my creativity and music making, at the end of the day I'm a musician, not a electronics engineer and a year later almost I've took a big detour into Neve neverland and haven't made much music since. I have gained a ton of knowledge though, witch is cool.
Enjoy your EZ unit's, they sound pretty badass when they are wired properly, and that's not to say they aren't wired properly how they come in the kit.... Again I'm only speaking on my experiences and conclusions after messing with all this for so long.
I've studied these units to the point where I'm sick of them. I know all the techniques and wired schems used in the originals. The grounding scheme is very specific in the original schematic. If you pay close attention you will start finding the differences in how the EZ1073 is wired up as well as the 2 resistors that are off from original value, I am surprised nobody has spotted those yet...
From what I've learned so far, the biggest impact on the sound of a 10XX class A module is the output transformer with the DC flowing through the primary and the gap at the core that is 0.003. The tolerances I believe are +/-20% with the Carnhill output transformer and probably even worse back in the day. I believe this is the reason allot of people say no two 1073s ever sound alike, even the same issue number and build date... It's that massive chunky LO1166, go back to what Rupert said in Sound City... that pretty much sums it up.
Don't let someone say it's components. Every single channel on the SSL G Series I work on sounds exactly the same and there's thousands of components on each channel. Capacitors and resistors are made to such strict specs today that those aren't to blame when things don't sound right. Transformers, are another story.
If you don't wire that output transformer correctly you will start getting stuff that doesn't sound right. It wasn't until I bypassed as much as the PCB as I could I started hearing the EZ 1073 units like I've never heard them before. If even 1 little tiny trace on the board is too close to another trace that it doesn't like, problems start happening. I'm not gonna say the EZ1073 board is flawed but I will say from my own personal experience with the 2 boards that were sent to me, they sound better with what I did to them, and clearly measure better too.
Even still at the end of the day, my AMS 1073 has something magic to it that sounds even better than my AMS 1084
(witch retails for a nice $4,295.00) Yeah $,4,295 for a preamp/eq... Anyways the AMS1073, it has a harmonic sparkle thing in the high end that even my 1084 doesn't do. My home etch units don't do it and neither do both of my EZ1073s.
All the same components in everything. Witch again really leads me to believe that those output transformers are one of the hardest things in the world to get perfect and even harder since a EM***** spec sheet was never made for the 1166. Leads me to believe my AMS 1073 has a really special output transformer in it by luck, and every single unit measures the exact same for freq and dist. This just makes even more sense to me when people say "something else is going on" It's just the variances in the output transformer we are hearing. The input transformer is pretty neutral. And come on, when you're paying AMS prices for preamps, don't you think they should all have the magic sound?
Go look at Avedis MA5 and you will start understanding why he has a custom Jensen output transformer made to his exact specs that he designed. He could have just used that Carnhill PCB output transformer...
Anyways, to sum it up. Experiment and more importantly get on to making music. This Neve quest thing has put a huge dent in my creativity and music making, at the end of the day I'm a musician, not a electronics engineer and a year later almost I've took a big detour into Neve neverland and haven't made much music since. I have gained a ton of knowledge though, witch is cool.
Enjoy your EZ unit's, they sound pretty badass when they are wired properly, and that's not to say they aren't wired properly how they come in the kit.... Again I'm only speaking on my experiences and conclusions after messing with all this for so long.
To be clear...I'm not listening for differences at this point...I'm measuring for differences. My curves are similar to what Aaron is getting, but what I'm trying to nail down is the unterminated response and bump past 20k.
As I was writing this post, I decided to take a few measurements and try my best to line those up. Doing so, I kinda have to walk back my previous statement about no difference. There is a difference in the repsonse and I suspect a pleasing one. However, no matter what I did, I could not get the unterminated response out at 30-50k to resemble what Aaron illustrated in a later post. That's were I was mainly looking earlier today.
Back to Aaron's work though...
The two areas I heard a difference in the AMS/home etch were in the "punchiness" of the sound and the "sheen" of the top. The EZ samples sounded "flat" and "midrangy" in comparison.
Looking at the graphs that Aaron has provided, part of it shows up. Look at his graph in post 240. Look at the phase response. There's a lift in phase response in the bottom, and in the top. That phase difference in the top could be responsible for the "sheen" aspect of the sound. How to get that? Shunt some highs to with a cap in the feedback loop? That kinda falls in line with Aaron's statements about the capactiance in the cable.
The next couple of posts I'm going to attach some graphs...
(graph removed due to inconsistent measurement)
As I was writing this post, I decided to take a few measurements and try my best to line those up. Doing so, I kinda have to walk back my previous statement about no difference. There is a difference in the repsonse and I suspect a pleasing one. However, no matter what I did, I could not get the unterminated response out at 30-50k to resemble what Aaron illustrated in a later post. That's were I was mainly looking earlier today.
Back to Aaron's work though...
The two areas I heard a difference in the AMS/home etch were in the "punchiness" of the sound and the "sheen" of the top. The EZ samples sounded "flat" and "midrangy" in comparison.
Looking at the graphs that Aaron has provided, part of it shows up. Look at his graph in post 240. Look at the phase response. There's a lift in phase response in the bottom, and in the top. That phase difference in the top could be responsible for the "sheen" aspect of the sound. How to get that? Shunt some highs to with a cap in the feedback loop? That kinda falls in line with Aaron's statements about the capactiance in the cable.
The next couple of posts I'm going to attach some graphs...
(graph removed due to inconsistent measurement)
Aaronrash
Well-known member
Hmm, what output transformers are you using? That is allot of ripple in the low end. The termination should kill allot of that but I see it isn't. Check to make sure there is no zobel on the PCB, that would cause a "Zobel loop" and cause all kinds of issues.
Also, post the distortion graphs too. Tell me your THD at 20kHz
Also, post the distortion graphs too. Tell me your THD at 20kHz
I'm using the "narrow board" Carnhill VTB1148. It appears to be physically and electrically identical to the VTB9049 which is what I understand to be the commonly used output.
The NV73 is a 500 format 1073 preamp. The zobel (as well as the termination resistor) is mounted on a PCB that mounts directly onto the transformer. I strip the PCB of all the components, but removing the PCB was looking pretty much impossible without destorying the PCB or the transformer. So the PCB on the transformer stays, but I wired the zobel and all connections directly to the pins on the transformer and not to the tiny holes on the PCB. The stock NV73 has a styrene in the zobel, I tried a PS series Sprague for giggles. Any film type cap shouldn't make too much of a difference here, and seeing as how the top end response is pretty much the same...I think that confirms it.
I haven't listened to the difference between the two units yet. But I think at this point there would be a slight edge to the modded one.
The NV73 is a 500 format 1073 preamp. The zobel (as well as the termination resistor) is mounted on a PCB that mounts directly onto the transformer. I strip the PCB of all the components, but removing the PCB was looking pretty much impossible without destorying the PCB or the transformer. So the PCB on the transformer stays, but I wired the zobel and all connections directly to the pins on the transformer and not to the tiny holes on the PCB. The stock NV73 has a styrene in the zobel, I tried a PS series Sprague for giggles. Any film type cap shouldn't make too much of a difference here, and seeing as how the top end response is pretty much the same...I think that confirms it.
I haven't listened to the difference between the two units yet. But I think at this point there would be a slight edge to the modded one.
Aaronrash
Well-known member
What is your THD at 20kHz as well as 20 HZ?
After seeing your graph, I think the AMS output transformers are indeed different. I've never had any low end ripple that drastic, even stranger that the 600R won't kill it... I really want to see the THD. That should tell us more about what's going on.
To look at THD, just look at the distortion tab for the same measurement and it will have a little graph, deselect all orders except for THD. when you sweep your mouse around it will read you your THD percentage in the corner.
After seeing your graph, I think the AMS output transformers are indeed different. I've never had any low end ripple that drastic, even stranger that the 600R won't kill it... I really want to see the THD. That should tell us more about what's going on.
To look at THD, just look at the distortion tab for the same measurement and it will have a little graph, deselect all orders except for THD. when you sweep your mouse around it will read you your THD percentage in the corner.
