ERC-1 Output Stage

[phyzar]

Hi there,

first post here after a lot of lurking
Thanks for this fantastic community, i´m already feeling like a member of the family, even if i´m thousands of miles away from the US. (germany)

I have a technical question to the output stage of the ERC-1.
I´ve seen two Burr Brown OPA2134 in there, but i´m not sure about the purpose.
Is this the I/V Stage or the drivers for the balanced outputs?


By the way, i found out that it is not a good idea to connect both unbalanced outputs simultanousely. (At least in my case)
I hooked Output 1 of the ERC-1 to a Linn Processor and Output 2 to the USP-1.
USP-1 was switched off, Linn was playing. Something was missing...imprecise bass, restrained mids.
I pulled out the connection to the USP-1 and "something was missing" turned into a big smile.
And no, it was not just the volume level that got naturally higher.
So give it a try, if you guys connected both unbalanced outputs.

Thanks in advance and greetings from germany

Wieland

[fishhead]

I've kind of wondered about those OPA1234's myself. It sure seems to be the opamps for the output stage. They are pretty cheap opamps and are dual channel. They go for about $1.65 at Digikey. I've wondered how well the ERC-1 could sound with four good quality single channel opamps. I wish someone with good electronic soldering skills would change them out for say AD797ANZ and let us know how it sounds.

I've been running the digital outs of the ERC-1 through a kit DAC with better grade opamps and getting a nice increase in SQ.

[SticknStones]

The Analog Devices was only $5-6 US on Avnet. I am not sure that price alone could differentiate which ones better.


Here is the Spec sheet on the AD product AD797ANZ
Description Value
Package 8PDIP
Type General Purpose Amp lifier
Typical Gain Bandwidth Product 450 MHz
Number of Channels per Chip 1
Typical Input Noise Voltage Density 1.7@±15V nV/rtHz
Typical NonInverting Input Noise Current Density 2@±15V pA/rtHz
Typical Slew Rate 20@±15V V/us
Typical Voltage Gain 146.02 dB
Maximum Input Bias Current 1.5@±5V uA
Power Supply Type Dual

---

Here is the TI Burr Brown OPA2134UA and there are 3 models in the 2134 family but I guess this is it. It sells for $1-1.5 on Avnet.

Description Value
Package 8SOIC
Type General Purpose Amp lifier
Typical Gain Bandwidth Product 8 MHz
Number of Channels per Chip 2
Typical Input Noise Voltage Density 8@±15V nV/rtHz
Typical NonInverting Input Noise Current Density 0.003@±15V pA/rtHz
Typical Slew Rate 20@±15V V/us
Typical Voltage Gain 120 dB
Maximum Input Bias Current 0.0001@±15V uA
Minimum CMRR 86 dB
Power Supply Type Dual

Maybe someone smarter than I can articulate the specs. ;D

[Mr. Ben]

Jan 25, 2010 14:16:15 GMT -5 SticknStones said:

The Analog Devices was only $5-6 US on Avnet. I am not sure that price alone could differentiate which ones better.

Keep in mind that the AD797ANZ chip is single channel, so you need two of them. The OPA2134 is dual-channel, so you only need one. I've been working on a DAC, and I believe I happen to have both chips around here somewhere. I just haven't directly compared them yet.

[phyzar]

Hi there,

at last some action in here ;D

Well fishhead, i can bring some sound-info in the near future, as i´m going to update the BB OPA2134 with Burson discrete Opamps.

Just to state it clearly: There´s loads of really good stuff in the ERC-1, this "tweak" is only for fun and it´s far away from "needed".

Examples? 2SK170 in the discrete output stage (Maybe a matching of these could(!) be another tweak)
Elna Caps, MPT(MKP?) coupling capacitors, 337 voltage regulator...just what i´ve seen at first look.

ERC-1 is an incredible value and purchase is indeed a no-brainer.

Well, i´m waiting for my Burson´s from down-under...i´ll keep you informed.

Wieland

P.S.: It´s the OPA2134PA, sticknstones

[SticknStones]

Jan 25, 2010 17:31:13 GMT -5 phyzar said:

Hi there,



P.S.: It´s the OPA2134PA, sticknstones

Oh Okay good clarification. Here are the high level specs from Avnet;

Description Value
Package 8PDIP
Type General Purpose Amp lifier
Typical Gain Bandwidth Product 8 MHz
Number of Channels per Chip 2
Typical Input Noise Voltage Density 8@±15V nV/rtHz
Typical NonInverting Input Noise Current Density 0.003@±15V pA/rtHz
Typical Slew Rate 20@±15V V/us
Typical Voltage Gain 120 dB
Maximum Input Bias Current 0.0001@±15V uA
Minimum CMRR 86 dB
Power Supply Type Dual

And a link where you can view the PDF files.

OPA2134PA Datasheet

AD797ANZ Datasheet

I suppose for those that have more of an EE background that the datasheets are informative. I did notice that there is a G=10 rating where both products are 8 MHz and a different G=1000 for the 450MHz rating for the Gain Bandwith spec but in all honesty this is outside of my particular area of expertise. If someone is so inclined to break it down then I am all eyes to read it. ;D

[Mike Ronesia]

Maybe you put the CD in upside down...

[fishhead]

No doubt that ther ERC-1 does contain many quality components, but I feel the OPA2134PA could be improved on. For one thing the OPA2134 is dual channel and most high quality ones are single channel. From what I can tell you need 2 singles for stereo output through RCAs and 4 singles for balanced outputs. Single channel opamps would surely keep the output signal more discrete. Also, from what I can derive reading about opamps(and some testing), the CMRR value seems to be one of the more important indications of the opamp performance.

From wikapedia: en.wikipedia.org/wiki/Common-mode_rejection_ratio
"In real applications, this is not always the case: the lower the CMRR, the larger the effect on the output signal. The 741, a common op-amp chip, has a CMRR of 90 dB, which is reasonable in most cases. A value of 70 dB may be adequate for applications which are insensitive to the effects on amplifier output; on the other hand, some high-end devices may use op-amps with a CMRR of 120 dB or more."

It seems the higher end chips are much quieter in their amplification of the signal for the output, and/or will allow a more complete signal to be amplified and outputed to the pre-amp. The CMRR of the OPA2134PA is rated at 86dB while the AD797ANZ is rated at 114dB. There is also an AD797BRZ rated at 120dB which got some of the highest ratings from some of the DIY sites.

I bought a DAC kit from China for $90 to do some playing with after reading some great reviews about it on the "Lamizator" site: www.lampizator.eu/LAMPIZATOR/LAMPUCERA/CD%20DAC%20Lampucera%20lampizator.html

Here is the DAC kit: cgi.ebay.com/24bit-192KHz-DAC-DIY-KIT-FULL-ASSEMBLED-KIT_W0QQitemZ230427368344QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item35a68a9b98

I then replaced the stock opamp on the kit( NE5532 dB70) with two AD797ANZ dB114, and hooked it up to the coax output of the ERC-1. I found that the AD797ANZ vs the NE5532 on the dac was noticeably quieter, more detailed, and gave better seperation and more complete and cleaner highs. I also found that outputting the ERC-1 through my DIY dac kit was giving me a darker, quieter sounstage with more seperation and better highs. The stock ERC-1 seems to roll off cymbals and bells, while this dac allows the full ringing sound to come through.

I know the ERC-1 has a good DAC chip, so possibly just changing out the opamp chips might make a nice increase in SQ. I'm really anxious to hear how the Burson opamps make it sound. Keep us posted.

[SticknStones]

Nice post and very informative! Thanks for sharing your research!

[RPA-1 man]

Interesting indeed.

[a2058]

Jan 25, 2010 17:31:13 GMT -5 phyzar said:

Hi there,

Well fishhead, i can bring some sound-info in the near future, as i´m going to update the BB OPA2134 with Burson discrete Opamps.
Well, i´m waiting for my Burson´s from down-under...i´ll keep you informed.

Wieland

P.S.: It´s the OPA2134PA, sticknstones

I am VERY interested in this mod as I have considered this path also. Please keep us posted, better yet, could you put up a review of the unit after the modification?

good luck!

[ntrain42]

I have a curiosity question: Why didn't they give the ERC-1 SACD playback ability since it uses the AD 1955 chip?

[lonnie]

Feb 17, 2010 10:48:37 GMT -5 ntrain42 said:

I have a curiosity question: Why didn't they give the ERC-1 SACD playback ability since it uses the AD 1955 chip?

Licensing. We didn't have a Macrovision (now Rovi) license at the time.

[ntrain42]

Feb 17, 2010 11:14:35 GMT -5 lonnie said:

Feb 17, 2010 10:48:37 GMT -5 ntrain42 said:

I have a curiosity question: Why didn't they give the ERC-1 SACD playback ability since it uses the AD 1955 chip?

Licensing. We didn't have a Macrovision (now Rovi) license at the time.

So will the ERC-1 be able to play SACD's with a firmware(assuming it can be updated)update or?

[Lonnie]

Feb 17, 2010 11:23:20 GMT -5 ntrain42 said:

Feb 17, 2010 11:14:35 GMT -5 lonnie said:

Licensing. We didn't have a Macrovision (now Rovi) license at the time.

So will the ERC-1 be able to play SACD's with a firmware(assuming it can be updated)update or?

No...the ERC wasn't intended for that, sorry.

[UT-Driven]

Feb 17, 2010 11:30:45 GMT -5 Lonnie said:

Feb 17, 2010 11:23:20 GMT -5 ntrain42 said:

So will the ERC-1 be able to play SACD's with a firmware(assuming it can be updated)update or?

No...the ERC wasn't intended for that, sorry.

Saving that for the XRC-1? ;-)

Doug

[ntrain42]

It would be nice if Emo was putting out another CD player that does SACD as well or something that competes a bit more directly in multiformat terms with the Oppo player which also plays blu-ray,DVD audio and SACD. Id be happy with just SACD though. Is there another player in the works from Emo?

[bigred7078]

Feb 17, 2010 13:27:18 GMT -5 ntrain42 said:

It would be nice if Emo was putting out another CD player that does SACD as well or something that competes a bit more directly in multiformat terms with the Oppo player which also plays blu-ray,DVD audio and SACD. Id be happy with just SACD though. Is there another player in the works from Emo?

there is always something in the works from Emotiva
Not sure how much sense it would make for them to put out just a CD/SACD player though. Like you said, a multi-format player would be a much more popular item

[hikinokie]

Something along the lines of a Perfect Wave Transport?

[solidstate]

Are the OPA2134 op-amps 8PDIP or 8SOIC ? I believe this question wasn't answered. I don't care if there is a socket because I could solder one in place if it's DIP.