The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
Help With Analog Delay Chips
Old 8th May 2013
  #1
Here for the gear
 

Help With Analog Delay Chips

I am really doing my research in trying to find the best analog delay that I can get for the best price I can get. The pedals I want the most are the Maxon AD999, The Diamond Memory Lane, The Delay Llama, MXR Carbon Copy, and EH Deluxe Memory Man. My problem is that I can see there are a lot of people out there marketing "analog delay" pedals, and when I research the chips they use, it turns out they are just digital "analog-replica" chips, which I by no means want. If I wanted an digital delay I would just buy a boss dd-6. Anyway, I am learning that what I need are the old "bucket brigade" style chips (and I have no idea what that means), or people have mentioned Panasonic chips. What I want to know is if there is anyone out there who really knows their stuff with these chips, because I am looking seriously at the possiblity of just having one built, but I would need to specify what chips I need. Also, I have seen a lot of people mention "dual-chip, or multi-chip" technology. Does it matter how many chips are in these pedals? Below are links of some of the custom models I'm looking at. Let me know what you think. Thanks in advance.

True MN3205 BB Analog Delay | 3pdt

Nocturnal Delay by NocturnalMechanics on Etsy
Old 8th May 2013
  #2
Lives for gear
 
JohnRoberts's Avatar
 

Well I know about these chips, since I designed several products using them 30 to 40 years ago, when they were new technology.

I would ask you why you absolutely positively must have analog delay, but i don't really want to know.

FWIW they fell out of favor in the 1980s, because digital had superior audio path performance, and now is cheaper too.

FWIW there may be some subtle characteristics in analog delay efx that digital does not routinely deliver, but it could. There is no magic involved, just different flavors of distortion.

JR
Old 8th May 2013
  #3
Moderator
 
narcoman's Avatar
 

The Carbon Copy and Maxon both have the chips you seek. So does the Aqua Puss.
Old 9th May 2013
  #4

The Delux Memory Boy uses the brigade chips too...

For a guitar pedal, the distortion is often what you want - especially as the regeneration fades.



-tINY

Old 9th May 2013
  #5
Lives for gear
 
Jazz Noise's Avatar
Quote:
Originally Posted by tINY View Post

The Delux Memory Boy uses the brigade chips too...

For a guitar pedal, the distortion is often what you want - especially as the regeneration fades.



-tINY

Nailed it in 1. It's the mucky, dark feedback. Often a touch of pitch modulation is added to make it "tape-y".

The Deluxe Memory Man w/ Hazarai is digital, but the rest of the Memory Boys are BBD types. They do a 1048 stage BBD now occasionaly when they get the chips - The Deluxe Memory Boy, it's a brown and green version. They also added extra expression pedal options, to my chagrin, I got the white and black and I quite like using an EP-2 for real time manipulation. The Send-Return path is cool too - stick a Crybaby in there!

What I really think alot of delay pedals fall down on is the use of a DPDT to swap the signal path rather than just removing the input from the delay. It feels abit underwhelming on stage when the delays just STOP.
Old 9th May 2013
  #6
Lives for gear
 
JohnRoberts's Avatar
 

IIRC there was a thread on this forum a while back (years) and a white paper published about mimicking the lossy transfer function of bucket-brigade charge coupled device analog shift registers, using digital technology. Perhaps too much work for a pedal.

JR
Old 9th May 2013
  #7
Quote:
Originally Posted by JohnRoberts View Post
IIRC there was a thread on this forum a while back (years) and a white paper published about mimicking the lossy transfer function of bucket-brigade charge coupled device analog shift registers, using digital technology. Perhaps too much work for a pedal.

JR

Actually, there are quite a few "analog modeling" digital delay pedals out there. They do pretty well for standard delay stuff - but they fall down in places like non-linear response to fast modulation. They don't seem to be programmed with all the warts....



-tINY

Old 9th May 2013
  #8
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by tINY View Post

Actually, there are quite a few "analog modeling" digital delay pedals out there. They do pretty well for standard delay stuff - but they fall down in places like non-linear response to fast modulation. They don't seem to be programmed with all the warts....



-tINY

Back in the early '70s I recall working with the then prototype ASRs from Phillips. Phillips later licensed the technology to Panasonic/Matshusita who sold a commercial version.

I recall a DC shift with clock frequency, that would add a changing voltage term to the audio signal from varying clock rates. If the clock is varied at an audio frequency rate, there are audible frequency artifacts, besides the obvious pitch shift from clocking in and out at different rates. In fact the application I was working on back in the 70's was using ASRs with a ramped clock for pitch shift correction to playback of pre-recorded talking books faster than real time with corrected normal pitch.

There are differential topologies using two ASR lines in parallel with audio inverted that generate a first order correction for that DC clock error term.

I guess I just can't fully embrace the flaws, after spending so much effort to reduce/remove them.

If you like BBD ASR distortion I should have saved some of those early puppies. They were really rough compared to later generations.

JR
Old 9th May 2013
  #9
Quote:
Originally Posted by JohnRoberts View Post
Back in the early '70s I recall working with the then prototype ASRs from Phillips. Phillips later licensed the technology to Panasonic/Matshusita who sold a commercial version.

I recall a DC shift with clock frequency, that would add a changing voltage term to the audio signal from varying clock rates. If the clock is varied at an audio frequency rate, there are audible frequency artifacts, besides the obvious pitch shift from clocking in and out at different rates. In fact the application I was working on back in the 70's was using ASRs with a ramped clock for pitch shift correction to playback of pre-recorded talking books faster than real time with corrected normal pitch.

There are differential topologies using two ASR lines in parallel with audio inverted that generate a first order correction for that DC clock error term.

I guess I just can't fully embrace the flaws, after spending so much effort to reduce/remove them.

If you like BBD ASR distortion I should have saved some of those early puppies. They were really rough compared to later generations.

JR

That's just the type of garbage that I look for in pedals for that "alternative" sound. It has to sound musical when dialed in though... tough to find just the right distortion sometimes...



-tINY

Old 9th May 2013
  #10
Gear Head
 

If you can find any, the SAD 4096 was an excellent delay chip.
Old 9th May 2013
  #11
Gear Maniac
 
Steele's Avatar
 

I just found some of those 4096 chips. The last delay I remember that used BBd was an old LOFT 450 delay. Before I could afford a Lexicon PCM-41.
Old 9th May 2013
  #12
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by StephenGiles View Post
If you can find any, the SAD 4096 was an excellent delay chip.
I don't recall the exact PN (5101?), but Reticon made a variant charge coupled device shift register that was actually much better performance than the bucket brigade devices but had an extra HF roll-off with clock frequency, but really nice dynamic range and clean sounding (which may be defeating the purpose of using these for some).

JR
Old 10th May 2013
  #13
Gear Head
 

I absolutely love my moog mf-104m.
Old 13th May 2013
  #14
Lives for gear
 

Quote:
Originally Posted by JohnRoberts View Post
I don't recall the exact PN (5101?), but Reticon made a variant charge coupled device shift register that was actually much better performance than the bucket brigade devices but had an extra HF roll-off with clock frequency, but really nice dynamic range and clean sounding (which may be defeating the purpose of using these for some).

JR
Yes, exactly, that was the Reticon R5101, one of their earlier parts and extremely hard to source now.
Old 13th May 2013
  #15
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by David Kulka View Post
Yes, exactly, that was the Reticon R5101, one of their earlier parts and extremely hard to source now.
As i recall it was developed after their popular BBD devices (SAD1024/4096/512d). While the BBDs are CCD (charge coupled devices) the CCD were a variant and not BBDs (IIRC 4 phases of clock inside not just 2). I had designed them into a consumer Hifi product, and Reticon stopped making the part a few years later. I was in the middle of a contract dispute with that client at the time. When the part went obsolete, all of a sudden they were my best friend, wanting me to redesign the product. I told them which body part they could kiss.

I don't know if Reticon lost the recipe or it wasn't profitable, or what, but it was a nice part. Since it was obsoleted in the '80s by Reticon, I expect they are scarcer than dragon's teeth now.

Probably too clean for a guitar pedal anyhow.

JR
Old 13th May 2013
  #16
Lives for gear
 
Richard Crowley's Avatar
 

Quote:
Originally Posted by JohnRoberts View Post
I don't know if Reticon lost the recipe or it wasn't profitable, or what, but it was a nice part. Since it was obsoleted in the '80s by Reticon, I expect they are scarcer than dragon's teeth now.
I suspect that Reticon may have had the chips made for them under contract by a "foundry" fab. If you are a "fabless" semiconductor company, you are at the mercy of your suppliers (the foundry fabs) to continue to provide the process that your chips was designed for. Processes change all the time as technology improves and they may have been shut-out because they couldn't find anybody to manufacture the chips anymore.

Or maybe they were simply made obsolete by digital technology which was growing at a remarkable rate in that era (as it it weren't today also). It is quite amazing that the things we used to with analog (or even mechanical) methods have been completely overtaken by digital solutions.
Old 13th May 2013
  #17
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by Richard Crowley View Post
I suspect that Reticon may have had the chips made for them under contract by a "foundry" fab. If you are a "fabless" semiconductor company, you are at the mercy of your suppliers (the foundry fabs) to continue to provide the process that your chips was designed for. Processes change all the time as technology improves and they may have been shut-out because they couldn't find anybody to manufacture the chips anymore.
Yup, Reticon was a small company, and had a surprising number of ICs, many performing obscure filter/transfer functions. I find it hard to believe there was much market for their other parts (perhaps academic and/or military?).
Quote:
Or maybe they were simply made obsolete by digital technology which was growing at a remarkable rate in that era (as it it weren't today also). It is quite amazing that the things we used to with analog (or even mechanical) methods have been completely overtaken by digital solutions.
I suspect this CCD was relatively popular in their catalog, but I can imagine having yield problems, running on somebody else's fab, where the fab owner wasn't sufficiently motivated to address the yield problem.

Yes, the writing on the wall was that digital technology and the still nascent DSP would moot pretty much their entire catalog,

I am intimately aware of the impact of digital technology on analog delay. I experienced it first hand.

JR
Old 13th May 2013
  #18
Lives for gear
 
Richard Crowley's Avatar
 

Yield is only one issue. An even more fundamental problem are the design details of the process itself. In the name of "progress" the foundry could have changed vendors for some material, or developed a "better" gate oxide or something that made everyone else's chips better, but cratered Reticon's design because of the unique requirements of a BBD circuit.

Here at the office (where we design the CPU chips used in most of the world's computers), we got out of the DRAM business decades ago and concentrated on digital/binary logic products. But then when customers started needing analog/linear capabilities (for high integration of mobile products), we had to revive the old analog techniques to create processes and devices for such arcane analog functions as capacitors and inductors along with our bleeding-edge digital/binary process and device engineering. Fortunately, design for digital clock cycles measured in GHz was very useful for implementing VHF RF functionality to support wireless functions like cell phones, wi-fi, Bluetooth, etc. etc.
Old 13th May 2013
  #19
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by Richard Crowley View Post
Yield is only one issue. An even more fundamental problem are the design details of the process itself. In the name of "progress" the foundry could have changed vendors for some material, or developed a "better" gate oxide or something that made everyone else's chips better, but cratered Reticon's design because of the unique requirements of a BBD circuit.
That would be a yield of 0%, but they are out of business well before yield gets that bad.
Quote:
Here at the office (where we design the CPU chips used in most of the world's computers), we got out of the DRAM business decades ago and concentrated on digital/binary logic products. But then when customers started needing analog/linear capabilities (for high integration of mobile products), we had to revive the old analog techniques to create processes and devices for such arcane analog functions as capacitors and inductors along with our bleeding-edge digital/binary process and device engineering. Fortunately, design for digital clock cycles measured in GHz was very useful for implementing VHF RF functionality to support wireless functions like cell phones, wi-fi, Bluetooth, etc. etc.
You get the clock frequency fast enough and it's all analog...

JR
Old 14th May 2013
  #20
Quote:
Originally Posted by JohnRoberts View Post
You get the clock frequency fast enough and it's all analog...

JR

...but THD+N = 30% is accepable....



-tINY

Old 14th May 2013
  #21
Gear Maniac
 
Berger's Avatar
 

Check out the malekko 616. I've owned an mxr carbon copy, an old plastic 80s mxr analog delay, a few digital delays, and the malekko is my favourite by far. Very dark sounding repeats. Sounds wonderful on synth or guitar. I've had it for 5 years or so and it's been gigged with a few hundred times with no issues.
Old 28th May 2013
  #22
A word on the MN3005 ICs from China on Ebay...

Run like hell !!!

I bought 2 batches of 10 from 2 sellers. All 20 were MN3008 2048 stage BBD and the Ebay pages say they are MN3005 4096 stage....

The printing on top of the IC says MN3005 but they are not, I tested them in a Electra EP150 analog delay box.
They were all counterfeit!!!

I am still fighting to get my money back less $47 I spent sending them back.

The only place I have had success with is Synth Cube, but that's here in the US.
Old 29th May 2013
  #23
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by RogerFoote View Post
A word on the MN3005 ICs from China on Ebay...

Run like hell !!!

I bought 2 batches of 10 from 2 sellers. All 20 were MN3008 2048 stage BBD and the Ebay pages say they are MN3005 4096 stage....

The printing on top of the IC says MN3005 but they are not, I tested them in a Electra EP150 analog delay box.
They were all counterfeit!!!

I am still fighting to get my money back less $47 I spent sending them back.

The only place I have had success with is Synth Cube, but that's here in the US.
Back in the '70s when I sold kits using these (actually earlier MN3001) Matshusita had distribution in the US and I was able to buy real panasonic parts from a real parts house. Pansonic was a quality IC company. These were premium parts and I don't recall even getting one dead on arrival, as was all too common from other IC companies back in the '70s.

I don't know if Panasonic even makes these still. I think some other IC company may make the MN parts these days.

caveat emptor.

JR
Old 29th May 2013
  #24
I suspect that all of these are the less popular MN3008, probably old stock that gets relabelled as the more desireable MN3005.

Seen 2 or 3 duds out of the 20 I tried to purchase.

I built your SAD4096 based delay unit in the late 70s / early 80s.
Then Craig Anderton put out a similar DIY project with enhanced modulation using the same IC. I stll have that unit in the garage and a spare SAD4096.

The thing I like are the really dirty/simple units that don't use telephony companders... True they are noisy as all get out, but the repeats are really interesting since the compander is not there to cut the repeats off early as they deteriorate into mud. Interesting mud.

When I want clean I use my Yamaha D500 or Roland SDE 330

I think I will put the Craig Anderton version in a proper rack case some day...

Roger
Old 29th May 2013
  #25
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by RogerFoote View Post
I suspect that all of these are the less popular MN3008, probably old stock that gets relabelled as the more desireable MN3005.

Seen 2 or 3 duds out of the 20 I tried to purchase.

I built your SAD4096 based delay unit in the late 70s / early 80s.
Then Craig Anderton put out a similar DIY project with enhanced modulation using the same IC. I stll have that unit in the garage and a spare SAD4096.

The thing I like are the really dirty/simple units that don't use telephony companders... True they are noisy as all get out, but the repeats are really interesting since the compander is not there to cut the repeats off early as they deteriorate into mud. Interesting mud.
If you connect the recirculation path from the output side of the compander to the input side of the compressor, the output expander won't alter the tail dynamics.

In addition to expanding the decay tail it would expand the narrow frequency dropouts due to comb filtering as the delayed signal interferes with the dry signal.

Another technique is to connect the side chain (rectifier) of the expander to the dry signal (output side of compressor) before the delay. The expander gain will lead the output slightly and open up a little sooner but in general this not audible and delivers stable dynamic tracking for short delays. Longer delays could experience audible artifacts leading loud transients.

I think I have even experimented with just running a buffered RC delayed version of the control voltage to the output side, while that doesn't un-distort the LF distortion from fast release in the compressor path.

I don't recall what I ended up with but there are numerous ways to skin that cat. Of course a little wide band noise enhances the sonic effect of resonant flanging.
Quote:
When I want clean I use my Yamaha D500 or Roland SDE 330

I think I will put the Craig Anderton version in a proper rack case some day...

Roger
There were some subtle things about my (LOFT) studio delay products that IMO were not a function of the dirty BBD path, while I am keeping myself busy enough these days pursuing objective design issues.

JR
Old 29th May 2013
  #26
Thanks JR, some good info there.. I never picked up a Loft 450, not for lack of desire, but lack of funds.
I bought the "Hi Fi" kit from Phoenix Systems (? I think)

There is not a very big market for this stuff, and the company Maxon, who made the venerable Ibanez AD202 is making their own BBDs now. Or possibly caving in to using shorter chips with their AD999... One second delays, but they are using like 8 BBD chips to get that. Have to be pretty short chips!

Elektra used 1- MN3005 for 300 ms delay at very restricted bandwidth.
The Ibanez is fairly Hi Fi comparatively.

Here I have 2- Ibanez AD202, 1- Elektra EP 150, 1- Maxon AD999, 1- Vintage MXR flanger and 1- Vintage ADA Flanger to round out the BBD boxes.
Back in the 80s I gave away my MXR green analog delay because I was after better fidelity... Still kicking myself!

I also bought a Yamaha E1010, but the idiot from ebay I bought it from put the unit in a 19" X 6" box with not internal protection. Got damaged obviously so I returned it. That guy really fought the return... That was an auction service company too (which will remain nameless) and you would think they of all people would use correct packaging, or not...

Now, IF I can get enough MN3005s, I am going to copy the Elektra EP150 and put 3 of them in a single units in series. That is why I am sourcing MN3005s.

Synth Cube is helping me out, but I am still looking for the MN3005s, so anybody gots some you know who to contact!
Old 10th June 2019
  #27
Here for the gear
 

Are you the John Roberts that wrote the articles in Popular Electronics in the mid-late 70's? If so, we have talked on the phone a couple times. You were very helpful and encouraging to a barely 21 year old tech. Your suggestion to "regenerate" the delayed flanger signal back to it's input was exactly what I needed. It took me a couple trys but I eventually figured out how to do that without it running away and squealing like a pig. I built probably a dozen or more of these over the years and they toured with some very respected rock bands. I just wanted to say "Thank You!" from the bottom of my heart, (no pun intended) for designing such a great sounding circuit, and for generously letting me run with it. btw - I have never found a BBD that worked as well for flanging as the MN3001. That chip had a great clock range and seemed to have a "sound" that I could not get from an SAD1024.
Old 10th June 2019
  #28
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by Rick Erickson View Post
Are you the John Roberts that wrote the articles in Popular Electronics in the mid-late 70's?
Yes..
Quote:
If so, we have talked on the phone a couple times. You were very helpful and encouraging to a barely 21 year old tech. Your suggestion to "regenerate" the delayed flanger signal back to it's input was exactly what I needed. It took me a couple trys but I eventually figured out how to do that without it running away and squealing like a pig. I built probably a dozen or more of these over the years and they toured with some very respected rock bands. I just wanted to say "Thank You!" from the bottom of my heart, (no pun intended) for designing such a great sounding circuit, and for generously letting me run with it. btw -
You are welcome.

My old kit designs were public domain (that was the deal with Popular Electronics). At least one of those old 1976 flanger kits ended up on a popular album, and probably more were used in performances.
Quote:
I have never found a BBD that worked as well for flanging as the MN3001. That chip had a great clock range and seemed to have a "sound" that I could not get from an SAD1024.
Back in the day I kind of liked the Reticon 512D for flanging. That IC had a clock driver built in (one less thing to screw up). The MN and SAD series used different polarity MOS technology, so technically the MN BBDs should be operated from a negative power supply, but rarely was. I didn't. You want to use a well regulated + supply so ripple on the positive rail doesn't inject hum into the audio path via the output source follower load resistors.

The old MN series made by Panasonic (actually Matsush-ita) was better quality/reliability than the Reticon stuff (in my experience). The MN stuff coming out of China these days is who knows? Process technology should have improved over the several decades since Panasonic stopped making them, but this may or may not translate to better audio performance.

JR

PS: I never was a fan of distortion and my sundry BBD designs were my attempt to deliver clean delay using the technology available. These days that is trivial using digital technology. In fact even early digital delay technology spanked BBDs for clean delay by the 80s.
Topic:
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