Help with Dolby schematic

SeleniumFalcon

I'm in need of some enhancement (no, not that kind of enhancement) concerning a Dolby B noise reduction unit, the Model 330. It uses the Dolby Cat. 66 module which is extremely difficult to find specific information about the schematic. There is only one schematic drawing and its resolution is low.



Is anybody aware of a program that allows for a file to be improved so that more can be seen? Thanks!

mhardy6647

So, @SeleniumFalcon I get the sense you've multiplexed several questions above!
Unfortunately, I can't answer any of them!
So, this NR system -- was it made by Dolby Labs (i.e., does it carry the "Dolby" brand name)?
Is it in fact Dolby B, and not Dolby A?
Are you looking for the module, a (better) schematic of the module, or a (better) schematic of the whole NR unit?
Or are you just looking for a way to "upsample" the image you have?

My one and only suggestion (which may be useless, i.e., you may have already explored this avenue): have you perused the searchable PDF database(s) at worldradiohistory.com? There are several recording and studio periodicals archived there. Seems like it would be worth a look(?).

Sorry I can't help more! Good luck in your quest!

mhardy6647

Ahh... this rascal?

SCompRacer

I've found obscure things at the Library of Congress but it only has the Dolby Cat 22 schematic.

https://www.loc.gov/static/programs/national-recording-preservation-plan/tools-and-resources/documents/Dolby_Cat_22_schematic.pdf

SCompRacer

Find Steve Koto? This 2020 thread said he modded them.

https://www.tapeheads.net/threads/modified-dolby-330-nr-unit.86159/

SeleniumFalcon

I have a Dolby 330 which was intended for cassette duplication and I'm trying to follow what Steve Koto's modification accomplished. So far I've done the standard power supply capacitor replacement and the output capacitor change. I've also rebuilt the Dolby level meter's indicator. With nothing done to the actual circuit it sounds better than my modified Advent/tube stage unit. I've been painstakingly using a magnifying glass to look at my laptop's screen to try and read some of the component values and part numbers. I've contacted Dolby Labs and they no longer offer product support. I've also contacted the owner of the site with the only existing schematic for a higher definition version, but unfortunately he no longer has access to the original manual. He said he has lots of manuals to look through so he might find it again.
I've also posted on a few professional recording studio websites asking for information, but no results. There was another person who tried contacting Steve Koto and received an initial response, but subsequently lost contact. The circuit board is fairly dense (as you can see) and I'm hand drawing the schematic by careful plodding along. I just thought there might be a computer program that could improve the resolution. This is a photo of what is probably a Steve Koto modified unit:



I've also tried to find information by using the patent number search, but that is a very complicated thread to follow.

SCompRacer

Do you have photoshop? That has a sharpening option IIRC. Unsure if it would provide desired results

Clipdat

Even MS Paint will let you zoom in on an image, but you can't restore/add resolution that isn't there to begin with.

SeleniumFalcon

"...but you can't restore/add resolution that isn't there to begin with."

That's what I was afraid of, I guess I was hoping for something like the scene in Blade Runner.

SeleniumFalcon

Here's a photo of the construction, there are 8 solid metal support braces that attach the upper and lower chassis plates. It's the strongest audio component I've ever seen. The basic power supply is on the left and the Cat. 66 module slides in place with channels made in 6 of the support columns. Normally there is a cover plate for the module that I've removed.

mhardy6647

FWIW, I sent a missive to a hifi colleague who's been in the business, so to speak, to see if he has any ideas.

SeleniumFalcon

Thank you, that's very kind, Mark. The Dolby Model 334 also uses the same Cat. 66 module. This was a noise reduction unit for FM radio broadcast.

txcoastal1

So none of these work??

https://www.bing.com/images/search?q=dolby B model 334 schematic diagram&qs=n&form=QBIR&sp=-1&pq=dolby b model 334 schematic diagram&sc=0-35&cvid=C06FAFC2E9A14DB4B3D3CE15F4E7B60C&ghsh=0&ghacc=0&first=1&tsc=ImageHoverTitle

SeleniumFalcon

Thanks I already have plenty of Dolby B schematics used by lots of other companies: Teac, Sony, Advent, KLH and more. I've collected 6 different Dolby decoders. Those are all commercial applications and have a few shortcuts in their designs.
The way a Dolby decoder works is by looking at the high frequency content of the input signal and reducing the high frequency content when there is no information there. This is how tape hiss is reduced by around 9 or 10dB. The way this is done is the incoming AC audio signal is converted to a DC control voltage that then varies the gain of an amplifier. The potential problem is the method this control DC voltage is obtained. To be less expensive in consumer equipment most decoders use a half wave detector (diodes) which only reacts to the positive portion of the music signal. Since music tends to be asymmetric when the high frequency response is being heard based only on the upper half of its waveform it will alter the way it sounds. The absolute phase of the music coming into the decoder will create an audible difference.
Since Dolby Labs wasn't under these budget extremes they can be more creative in how this control DC voltage can be generated. That's why unlocking the schematic is so necessary.

txcoastal1

Very interesting but a bit discouraging that a decent schematic cannot be found for that model with all the tools we have now days.
I googled around and you almost get there but not quite.

So to confirm, this would be in front of the tape player while recording. Would this also work on the output side during playback?

Good luck

Emlyn

There are applications that can use AI to approximate blurred letters and numbers from an image using sharpening algorithms. Focus Magic and Adobe Express are a couple. Focus Magic costs money but Adobe Express is free to use.

I haven't tried this myself.

SeleniumFalcon

Thanks! Yes, there's a front panel toggle switch for "encode" or "decode". I would have thought Dolby Labs would, at least, have the owner's manuals available.

Clipdat

They probably do have it archived somewhere, but there's no incentive for them to release it.

CoolJazz

Maybe the last page of this link will be clearer reading.

https://bh.hallikainen.org/uploads/HaroldHallikainen/Dolby334.pdf

Good luck!

CJ

SeleniumFalcon

Thanks, @CoolJazz, I appreciate the link. I posted this search request on an Ampex group site and a very obliging member in Switzerland sent me an email with the schematic very well reproduced. I was finally able to determine part numbers and values. The first thing I wanted to figure out is the very steep input filter, which seemed to be between 18kHz and 20kHz. Since this model was intended to be used in cassette tape duplication there wasn't much need for frequencies above 17 or 18kHz. I began by removing one of the filter capacitors and installing two small wires that I could try different value capacitors (around 2.0nF to 3.0nF) and see what the response became. I found I could reduce the roll off from -10dB at 20kHz to -6.4dB by substituting different values. Then I noticed that there was the ability to move some jumper wires on the edge card socket and began looking at the schematic closer and tried changing the jumper positions and found one that gave me what I was looking for. The frequency response is within 1.5dB from 20 to 20kHz. I've done both channels and plan to take a listen tomorrow after I let the unit warm up over night.
Now to change parts in the regulated power supply section and a few coupling caps.

txcoastal1

That's great news. Kinda funny about the jumpers (simple if you know what to look for ) otherwise perplexing

YaY!! to the Swiss...congrats and hope she gives you what you're looking for

SeleniumFalcon

The key was figuring out how the parts listed on the schematic connected to the edge card connector now that I could finally read them. I started by finding a capacitor on the board then traced it to where it went to the connector. After tracing that for a few components I could tell where A8 mated with the connector and how to number the connections for the upper and lower set of pins. I initially thought that I could push the low pass filter point higher by changing one of the capacitor values. The filter uses two variable coils and series and parallel capacitors to make the filter.



After ordering some additional small value capacitors and experimenting with changing C104 I believed I had reduced the effects of the filter as much as I could, at that point. I tend to think about problems I'm trying to solve just as I wake up in the morning, funny habit. I could see the circuit in my mind and hit on the section that showed three leads with alphanumeric labels, but no switches or selectors connected, just the small letter/numbers. Then it dawned on me that there's another chart that shows what I thought was the multi-pin connector and the leads attached from inputs, outputs power supply connections, etc. One of them showed a jumper link and I could see where the chart corresponded to the three labels on the schematic. A ta da moment.

SeleniumFalcon

I was able to use my Bruel and Kjaer level recorder to make some very specific frequency response measurements of the unit. I had to make the pen holder myself from Starbuck's stir sticks and wood glue, on the third try I got a working version.

SCompRacer

Very clever fix with the pen holder! Wish you lived nearby....

C104, 3900pf, what value did you end up with?

OleBoot

Kenneth MacGyver.

SeleniumFalcon

"C104, 3900pf, what value did you end up with?"

I ended up joining points B8 and B9 which makes the circuit path: R105, inductors L100 and L101 in series (the inductances add together) and C106 (2200pF) in series to ground. Both channels are down 1.6dB at 20kHz and fall off steeply after that. From reading how the Dolby process works there has to be a steep filter to eliminate any problems with the compression/expansion circuits that happen in the next stage.

SeleniumFalcon

SeleniumFalcon

A collection of Barclay Crocker tapes up for auction on eBay:

mhardy6647

yowsa!

SeleniumFalcon

According to the eBay advert the collection is only missing four titles, plus they have an unopened Dolby calibration tape and some special subscription-only tapes of Christmas gifts from B-C and a decoder is included. The opening bid is 10k or "best offer", not sure what that means.

pitdogg2

https://www.ebay.com/sch/i.html?_dkr=1&iconV2Request=true&_blrs=recall_filtering&_ssn=acousticsoundsinc&store_cat=0&store_name=acousticsoundsinc&_oac=1&_nkw=reel to reel music

https://store.acousticsounds.com/index.cfm?get=topsellers&Top_Count=200&field_cat=397


SO FAR OUT OF MY LEAGUE!!!