The No.1 Website for Pro Audio
Help Identifying Type of Film Cap
Old 31st January 2020
  #1
Lives for gear
 
Brian M. Boykin's Avatar
Help Identifying Type of Film Cap

These are in the filter section of an Audioarts 4100 parametric EQ I’m recapping. If they’re Mylar then I’d like to replace them.
Attached Thumbnails
Help Identifying Type of Film Cap-7524a9a7-30b6-4085-8e95-6008cd14a1ce.jpg   Help Identifying Type of Film Cap-a8ea48af-1629-4165-a1b6-6b0dcc68e1f7.jpg  
Old 31st January 2020
  #2
Lives for gear
If you change them you will likely change the characteristic sound of that electric crayon. Those boxes have such a distinct character... maybe if you are looking for something super clean and transparent, you should check out the Orban 642.
Old 31st January 2020
  #3
Lives for gear
 
Brian M. Boykin's Avatar
Quote:
Originally Posted by enginefire View Post
If you change them you will likely change the characteristic sound of that electric crayon. Those boxes have such a distinct character... maybe if you are looking for something super clean and transparent, you should check out the Orban 642.
I would venture to say the RC4136 op amps give it its distinguishable slurred veiled sound over the caps. I replaced those in all my Ashly SC50’s and 55’s and they became much more transparent, clean and usable. Which is what I’m looking for. I’m not looking for a character piece as much as a tool I can use for small tweaks without completely changing the original source material. I have two of these and a stereo version. I may have $300 total invested in the actual EQ’s. $200 in parts and my time and I should have 3 very usable EQ’s that will work across any platform.

Do you know if the caps in the photo’s are Mylar? I can’t seem to google and find a reliable source on their type.
Old 31st January 2020
  #4
Lives for gear
The ones in the second photo are polystyrene, those are what I would use personally, they are better than any polypropylene imho. The orange ones... might be Mylar, but look to me more like polypropylene in a style like orange drops. Their large size and shape makes me think this. Can’t confirm though.

The 4136... it’s aN old TI so it probably sounds decent. There is also a 5534, not a shabby amp at all. Without looking at the schematic (do you have it? Can you post it please? I’d love to see it).

Maybe you can make your mod process a little scientific? Some objective before and after samples for A/B comparison?

I suspect that you will get the results you are after by changing all the opamps. They’re socketed, why not give it a quick try and see?

Personally, I stand by the idea that if a well adored character eq is not what you are looking for, sell it to someone who is looking for that and buy something that already does what you want. There are lots of options for clean and surgical.

But if quick minor tweaks are what you need, that are clean and transparent, why not just use a plugin?

Maybe I’m biased, I for one like these EQs for their character. They sound great and often are just what a track needs. You can do things with them that no other eq can do.
Old 31st January 2020
  #5
Lives for gear
https://www.google.com/search?q=vint...g4QP-52i11HHjM

Looks like they are metalized polyester film aka MKT caps.
Old 31st January 2020
  #6
Lives for gear
 
Brian M. Boykin's Avatar
Polystyrene was my guess for the silver ones and I that case I don’t want to replace them. If the Nichicons are MKT’s then I’ll swap them for MKP’s.

I don’t have the schematics. I’ve unwound the 5534’s in the circuit. They all have feedback loop caps and I added local psu caps from the +- pins to ground. None of the 4136’s have feedback caps. They also got local psu caps added. They’re all in sockets. I’m generally going for lower noise and distortion so I swap to lm4562’s for bipolars. Can’t get those in a single and I don’t trust NOS lme49710’s anymore. I’m thinking opa1611’s on browndog adapters. I have some 4136 adapters loaded with lm4562’s on order for another project so I’m gonna try them and see how they like circuit. The noninverting pin is grounded on all the audio op amps except the input stage of the balanced input. No resistor between the pin and ground. I have not broke down the feedback loops yet of the 4136’s. Schematics would help but I haven’t been able to find them.
Old 13th February 2020
  #7
Lives for gear
 
Brian M. Boykin's Avatar
Finished the first one. Sounds great. I changed the Nichicon’s to WIMA’s, the 4136’s to LM4562’s on browndog adapters, and the 5534’s to opa1611’s on browndog adapters. Each op amp got local psu decoupling caps from rail to ground and the 5534’s that didn’t have feedback caps had them added.

The second one had major distortion issues. Turned out to be the +15v rail was down 3v. Took me a while to find the reason but it cleaned right up once I did. It’s also missing some resistors in the filter sections. Not sure why. Sonically it sounds the same. The PCB’s are identical and the traces and pads are there so I’m gonna add them. Next is to put all metal film in the audio path.
Old 13th February 2020
  #8
Lives for gear
Post some sound files please! I would love to hear what they sound like with these changes.
Old 13th February 2020
  #9
Lives for gear
 
Brian M. Boykin's Avatar
Quote:
Originally Posted by enginefire View Post
Post some sound files please! I would love to hear what they sound like with these changes.
Give me a week so I can order all the op amps for the second one. I didn’t want to commit to $130 in parts for the second one until after I heard the first. These will be going on the mix bus post compression. They’re replacing two KT DN27a’s. You have some reference songs you’d like me to use? I typically use EDM because it so clean and it’s easy to hear what’s being done.
Old 13th February 2020
  #10
Lives for gear
Anything is good, but if you are willing to outside your comfort zone anything from the late 50s eta of miles Davis would be amazing. Anything acoustic as in non electronic, I think that would contrast well with the EDM to give a full picture!

Are you mixing or mastering with this chain? Which compressor are you using on the 2bus before the eq? Sorry just geeking out a bit
Old 13th February 2020
  #11
Lives for gear
 
Brian M. Boykin's Avatar
Quote:
Originally Posted by enginefire View Post
Anything is good, but if you are willing to outside your comfort zone anything from the late 50s eta of miles Davis would be amazing. Anything acoustic as in non electronic, I think that would contrast well with the EDM to give a full picture!

Are you mixing or mastering with this chain? Which compressor are you using on the 2bus before the eq? Sorry just geeking out a bit
I mix OTB. Compressors are Aphex 651’s. I have 4 of them. Two for vocals which Jim Williams modded, and the two on the mix bus which have not been to see him yet. I’m going for the lowest noise, lowest distortion, and greatest transparency in these rolls. Thus my choice in parts. Wheatstone emailed me the schematics so I’m gonna try direct coupling as much of the EQ’s as possible. I’ve never tried before so I’m a little timid. At the least I’ll bypass the electro’s in the audio path with films.

I also have a 4200 which I’ll be giving the same treatment to. It will replace a KT DN22 I use on drum OH’s. I’m getting the inductors out of the chain for now. The only inductor EQ’s left will be the White 4200a’s.
Old 13th February 2020
  #12
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by Brian M. Boykin View Post
These are in the filter section of an Audioarts 4100 parametric EQ I’m recapping. If they’re Mylar then I’d like to replace them.
The clear/silvery looking caps are polystyrene, a very good dielectric, you will not improve upon.

The orange ones are film, perhaps polyester (Mylar?), maybe polypropylene.... I can't tell from the picture. Probably not the weakest link.

The 4136 is not up to modern op amp standards but was popular back in the day. 1.7V/uSec is not blazingly fast, but unlikely to slew limit for typical audio signals.

JR
Old 13th February 2020
  #13
Lives for gear
 
Brian M. Boykin's Avatar
Quote:
Originally Posted by JohnRoberts View Post
The clear/silvery looking caps are polystyrene, a very good dielectric, you will not improve upon.

The orange ones are film, perhaps polyester (Mylar?), maybe polypropylene.... I can't tell from the picture. Probably not the weakest link.

The 4136 is not up to modern op amp standards but was popular back in the day. 1.7V/uSec is not blazingly fast, but unlikely to slew limit for typical audio signals.

JR
Thanks John. I left the polystyrene because that’s what I concluded also. I replaced the films because I couldn’t verify what they were using Google and they’re so cheap it was worth it to me. All the 4136’s were swapped for lm4562’s on browndog adapters and the 5534’s for opa1611’s. Missing supporting circuitry added. All electro’s recapped value for value except for a small bump in the filter caps.

Did you ever test or have you read how much nV of noise it takes for the human ear to be able to tell a difference? Id be interested in any info you have or could direct me to.
Old 13th February 2020
  #14
OPA1602's are the same as LME49720/LM4562 with the BurrBrown rail to rail output. SOIC only, 2.5 nv noise. OPA1612's are 1.1 nv noise. ADA4898-2 is .9 nv noise.

Resistor thermal noise will add to the results. Any resistors over 10k ohms may show less overall noise than the 1611 if you use fet input opamps like the OPA1652.
Old 13th February 2020
  #15
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by Brian M. Boykin View Post
Thanks John. I left the polystyrene because that’s what I concluded also. I replaced the films because I couldn’t verify what they were using Google and they’re so cheap it was worth it to me. All the 4136’s were swapped for lm4562’s on browndog adapters and the 5534’s for opa1611’s. Missing supporting circuitry added. All electro’s recapped value for value except for a small bump in the filter caps.

Did you ever test or have you read how much nV of noise it takes for the human ear to be able to tell a difference? Id be interested in any info you have or could direct me to.
It isn't quite that simple but if you have too much time and too much money lower is considered better. I ASSume you are referring to nV per root Hz, an equivalent input noise (ein) voltage per unit bandwidth. That number gets multiplied by 20k or so to look like a recognizable voltage, but that is just one term of several when evaluating circuit noise floors. Besides the ein voltage, there is also an equivalent input noise current that must get multiplied by the effective input resistance to estimate a second noise term that must get combined with first (not added but square root of sum of terms squared). Then there is the self noise (aka Johnson or thermal noise) of the resistance itself. Then that total EIN gets multiplied by the noise gain of the circuit.

Its been several decades since I looked under the hood of a Gary Snow(?) design. IIRC he favored state variable filter (SVF) topology for parametric EQs which further complicates your noise analysis since the sundry different op amps stages have different noise gains and bandwidths.

I do not mean to scare you with science... you might save a few bucks by not throwing uber fast opamps at the LF and mid band passes of the SVF.

Noise gain depends on the specific topology and in general lower noise op amp will deliver lower noise.

You could experiment with one band of EQ at a time and see how much improvement you can realize. The 5534 is still respectable today if properly applied, 4136 not so much, perhaps as evidenced by the fact they still make 5534s.

JR
Old 13th February 2020
  #16
Lives for gear
 
Brian M. Boykin's Avatar
Quote:
Originally Posted by Jim Williams View Post
OPA1602's are the same as LME49720/LM4562 with the BurrBrown rail to rail output. SOIC only, 2.5 nv noise. OPA1612's are 1.1 nv noise. ADA4898-2 is .9 nv noise.

Resistor thermal noise will add to the results. Any resistors over 10k ohms may show less overall noise than the 1611 if you use fet input opamps like the OPA1652.
Thanks once again Jim. I had the noise specs on the 1612’s but the cost from Cimmaron was steep compared to the trusty LM4562. I’m gonna try my hand at dropping some ada4625-2’s on some browndog adapters per your recommendation for the Ashly SC63 I went through. If I’m successful it opens a door to an entirely new level of op amps and will lower the cost also.

Are you saying it’s better to switch to FET input op amps if the resistor is over 10k? Or is that too general? I’ve been revisiting old mods and dropping in bipolars for bipolars and FET’s for FET’s where I mixed and matched.
Old 13th February 2020
  #17
Lives for gear
 
Brian M. Boykin's Avatar
Quote:
Originally Posted by JohnRoberts View Post
It isn't quite that simple but if you have too much time and too much money lower is considered better. I ASSume you are referring to nV per root Hz, an equivalent input noise (ein) voltage per unit bandwidth. That number gets multiplied by 20k or so to look like a recognizable voltage, but that is just one term of several when evaluating circuit noise floors. Besides the ein voltage, there is also an equivalent input noise current that must get multiplied by the effective input resistance to estimate a second noise term that must get combined with first (not added but square root of sum of terms squared). Then there is the self noise (aka Johnson or thermal noise) of the resistance itself. Then that total EIN gets multiplied by the noise gain of the circuit.

Its been several decades since I looked under the hood of a Gary Snow(?) design. IIRC he favored state variable filter (SVF) topology for parametric EQs which further complicates your noise analysis since the sundry different op amps stages have different noise gains and bandwidths.

I do not mean to scare you with science... you might save a few bucks by not throwing uber fast opamps at the LF and mid band passes of the SVF.

Noise gain depends on the specific topology and in general lower noise op amp will deliver lower noise.

You could experiment with one band of EQ at a time and see how much improvement you can realize. The 5534 is still respectable today if properly applied, 4136 not so much, perhaps as evidenced by the fact they still make 5534s.

JR
Thanks for the in depth answer. Your assumption about the noise spec is correct. I didn’t throw in the square root sign. Basically your saying the number is so small we multiply so we have something we can make sense of.

The reason I do this is because money doesn’t grow on trees around here. I’ve found it much more financially feasible to buy a used piece of gear that’s ripe for mods and spend the time with a soldering iron putting modern low noise parts in them rather than spending money on new designs that don’t offer much better performance. I know it’s more complicated than that but It has served me well.

I do most of my mods in stages. The recap always makes the biggest difference sonically. Op amps seem to refine the image if that makes sense? That’s up for argument however. As you say, “I can’t hear what I can’t measure.” I have roughly $300 invested in this EQ. Not bad and I’m learning. That’s priceless.
Old 13th February 2020
  #18
Lives for gear
 
JohnRoberts's Avatar
 

Not to bury you under too much information, you mentioned FET input op amps. The classic trade off there is that FET op amps generally have higher ein noise voltage but orders of magnitude less noise current, so indeed there are impedances where FET can be quieter than bipolar with ALL noise sources considered. Of course when the resistance gets large enough it's self noise can be significant.

JR
Old 13th February 2020
  #19
Lives for gear
 
Brian M. Boykin's Avatar
Quote:
Originally Posted by JohnRoberts View Post
Not to bury you under too much information, you mentioned FET input op amps. The classic trade off there is that FET op amps generally have higher ein noise voltage but orders of magnitude less noise current, so indeed there are impedances where FET can be quieter than bipolar with ALL noise sources considered. Of course when the resistance gets large enough it's self noise can be significant.

JR
Bury me. I don’t always grasp it right away but usually I learn the missing link and then it all makes sense. I know circuit topography matters. Lightbulbs go off and I go back and reread threads where you brains are hashing it out. There are classic discussions hidden in this forum that I reference and reread. The learning never ends.
Old 14th February 2020
  #20
Lives for gear
Quote:
Originally Posted by Brian M. Boykin View Post
I mix OTB. Compressors are Aphex 651’s. I have 4 of them. Two for vocals which Jim Williams modded, and the two on the mix bus which have not been to see him yet. I’m going for the lowest noise, lowest distortion, and greatest transparency in these rolls. Thus my choice in parts. Wheatstone emailed me the schematics so I’m gonna try direct coupling as much of the EQ’s as possible. I’ve never tried before so I’m a little timid. At the least I’ll bypass the electro’s in the audio path with films.

I also have a 4200 which I’ll be giving the same treatment to. It will replace a KT DN22 I use on drum OH’s. I’m getting the inductors out of the chain for now. The only inductor EQ’s left will be the White 4200a’s.
I’ve heard those are nice compressors but have never tried them. Clean and transparent are great qualities!

Could you upload the schematic please? One of my units has a band that does nothing, one day I will have to open it up and take a look. I’m curious now to see what caps they’re using and if either of my units are missing parts!

Great thread.
Old 14th February 2020
  #21
Lives for gear
 
Brian M. Boykin's Avatar
The 651’s are the best of that line. They fly under the radar and can be had for $100 to $150 and if your patient I paid $180 for two delivered to my front door. The biggest complaint people have is you can’t hear them doing their job. If you want transparent then look no further. After Jim does his thing they’re just plain stupid. A seller was unloading 20+ units a while back at $60 a pop. I should have bought more.

Anyway, here are the schematics. They sent me the 4200 schematics also and they’re identical. I haven’t popped mine open but I’m guessing it’s two 4100 PCB’s on top of one another.
Attached Files
File Type: pdf AE model 4100.pdf (2.35 MB, 9 views)
Old 14th February 2020
  #22
Lives for gear
Quote:
Originally Posted by Brian M. Boykin View Post
The 651’s are the best of that line. They fly under the radar and can be had for $100 to $150 and if your patient I paid $180 for two delivered to my front door. The biggest complaint people have is you can’t hear them doing their job. If you want transparent then look no further. After Jim does his thing they’re just plain stupid. A seller was unloading 20+ units a while back at $60 a pop. I should have bought more.

Anyway, here are the schematics. They sent me the 4200 schematics also and they’re identical. I haven’t popped mine open but I’m guessing it’s two 4100 PCB’s on top of one another.
Thanks this is super helpful; I spent way too long searching for this and all the while it was only an email away!

I’ll keep my eyes open for the 651s!
Old 14th February 2020
  #23
Lives for gear
 
Brian M. Boykin's Avatar
The 4.7Mohm resistors in the filter section is what’s missing on my second one. The 3 5534’s that show feedback caps of 0 have physical spots on the PCB for the caps but they were not there. I added them. I had 33pf’s on hand so that’s what I placed and I swapped in 33pf’s for all the 5pf’s for consistency. WIMA does not make a .18 film so I chose to go up in value to .22. Everything else was value for value. Looking at the first stage of the filter section it has a 10k resistor across the feedback loop with no cap. I’ll put a 33pf there for added security. There are lots of decoupling caps in this design. I would think a good number of the 10uF’s could come out and maybe some of the 220uF’s.
Old 14th February 2020
  #24
Quote:
Originally Posted by Brian M. Boykin View Post
Are you saying it’s better to switch to FET input op amps if the resistor is over 10k? Or is that too general? I’ve been revisiting old mods and dropping in bipolars for bipolars and FET’s for FET’s where I mixed and matched.
It should be examined and compared. Modern fet input opamps are now at the low noise levels as their bipolar counterparts, this is a major development in opamp manufacturing. 2.9 nv noise for a fet device was unheard of just a couple of years ago. The LM4562 is 2.7 nv. The resulting noise can be less with the fet device if resistors are larger than 10k ohms.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump