Reverend Lucas, post: 420739, member: 48050 wrote: I like bad puns and am all about shameless self-promotion

That deserved a Like!

Reverend Lucas, post: 420739, member: 48050 wrote: Has anybody heard of anyone building such a thing? Should I just be building one of each Rev? Can I quit my day job and start building these? Have I not had enough caffeine today?

The answers to your questions are:

1. Yes. There are guys out there who build their own versions of classic studio processors.

2. It all depends on how good you are at this type of thing. If you can knock out one of each in a day or so, then that leads us to the next answer...

3. Yes. Most definitely... Rev.. Rev... time to wake up now.. it's time for school... c'mon now, sleepy head... wake up)

4. It sounds like you have. Perhaps cut back to a half-caf blend.

Seriously.... of course you can do this, if you can in fact do this. If you have the knowledge, the means, the materials, the tools, then why not? There are individuals out there who do make clones of classic pieces for sale, and they are a viable alternative as long as they:
A. do what they are supposed to do
B. are built well
C. priced reasonably

Start out by making one for yourself, and see how it goes. Keep track of your costs (including your time!) and then, if you find that you can make one or two in a reasonable amount of time, send a few out to some beta testers - guys like Chris, Kurt, PC, or , ahem, Me - LOL - or others who have experience with the originals, and who can then testify to the quality of the pieces.

Don't get greedy. You'll be competing with some big names, some of whom are remnants of the original companies who made these models to begin with. Be fair to yourself in terms of profit, but don't mark it up to the point where people get scared away, or who might think that because your price is so close to that of an actual reissue from one of the original manufacturers, that they might as well just get one of those instead.

At that point, you start your marketing, take deposits from customers, and you're on your way.... hopefully. ;)

Thanks for the encouraging and realistic words, Donny. You'll be the first to get a proto if this comes to light.

I have a degree in electrical engineering and a modestly equipped solder bench at home, but lack in the practical side of things.

I'll maybe start by building one, and see how it goes. The trouble is all the custom stuff is so much cheaper in quantities. I think the trick is getting efficient enough with the materials and build to break even, or maybe even eke out minimum wage. I wouldn't expect to be getting rich off of them.

Chris, I thought of another bad pun you might enjoy if you liked the first. The switch between the rev circuits will be called the A to D converter. (Feel free to groan)

I'm still looking to see if anyone's built both the Rev A and the Rev D into the same chassis. Haven't found one yet.

Reverend Lucas, post: 420739, member: 48050 wrote: .. I haven’t looked into the two circuits enough to know how feasible this is. It seems to me the most expensive components (input/output transformers, power supply, VU meter) could be shared with some switching that’s hopefully not too complicated.

If you could point to any links to the schematics of both the A and D revisions, I'm willing to take a look at them and give an opinion on whether they could share the same box. Alternatively, if you have scans of the schematics, PM them to me via a Forum Conversation.

I know the A version used FETs in the signal pre-amp that were replaced by 2N3391A transistors at revision B and again by a separate low-noise circuit (the LN in the model number) at revision C. This was incorporated on to the main board at revision D. There were changes to the output transformer and drive circuitry as well, but it could be that those came in at revision F.

Did you have any thoughts about stereo linking of a pair of these compressors?

Boswell,
I'd definitely appreciate your expert opinion. I've been looking at the mnats schematics, which a lot of DIY builds use. Without having looked at the original schematics, they claim to be fairly faithful to the originals, with minor concessions to replace now hard-to-find components.

Rev A:
http://mnats.net/files/1176REVA_125_DOCUMENTATION.pdf

Rev D:
http://mnats.net/files/1176LN_REVD_V2.2_DOC.pdf

The biggest differences are like you said, Boswell, replacing FETs in the preamp and output amps with BJTs. The power supplies and input/output transformers are identical, with minor differences in the metering and gain reduction circuits.

It looks like I have some homework to do. Part of me thinks that someone would have done this if it were easy to do.

Stereo linking capability would be a great thing. Of course, then I'd have to have two. I wouldn't be horribly sad about that.

Thanks for linking the two schematics. I've had a quick glance at them, and the idea of building a Rev A/D (call it a Rev L?) switchable box is not all that far-fetched. The real difficulties would be more physical ones such as how you would get two parallel paths of input and line amp circuitry in the one original-style box.

If you were starting with your own new PCB layout, things would be relatively straightforward, but much more tricky if trying to use the mnats PCBs. A compromise might be to layout the Rev A signal path from a point at the secondary of the input transformer to just before the emitter follower in the line amp along with the necessary provision for signal switching on a new small PCB and then devise a way of mounting that above the main PCB in what was an otherwise Rev D box. I suggest that way round because of the fewer components in the Rev A signal path.

While doing this, it may be possible to incorporate a version of the Purple [[url=http://[/URL]="http://mnats.net/fi…"]stereo link board[/]="http://mnats.net/fi…"]stereo link board[/] components so that future stereo operation would be just a matter of connecting a similar box. It would probably mean sorting through a big bag of 2N5457 FETs to pick out a couple that were closely matched and saving one of them for your second box build.

Thanks again for the insight, Boswell.

I like your approach. It looks to get to the core differences between the two (preamp and lineamp), without unnecessarily duplicating parts. I would imagine the differences in the gain reduction/metering circuits have comparatively minor impact to the sound.

A stereo link is a few dollars' investment that I might be very glad about later. Foresight!

A custom PCB would be the way to go. I quickly quoted a couple places today, and the tooling for an order was $400-$500. That's a bit much for a prototype, but might be worth it if the build goes well and there's interest in them. I'm trying not to get ahead of myself...

Reverend Lucas, post: 420784, member: 48050 wrote: Part of me thinks that someone would have done this if it were easy to do.

I find fault with that statement. ;)

Don't let yourself think that. First of all, it may not be true. Have confidence in yourself and accept the possibility that the ideas you've come up with could be very good ones. Second, consider the possibility that "no one has done it yet" does not mean that it's not do-able.

It may in fact be very difficult. But that shouldn't stop you, either.

If you really want to do this, you should.

Oh...and one more thing....You're in mighty good hands with Boswell. To say that he is highly intelligent would be an understatement. ;)

I've got some grand-child version of the 1176 in my UA 4-710. Having a, stereo, closer to the original version would be fantastic (if affordable)

You're not throwing 50k to develop a product here. it's worth a try ;)

Reverend Lucas, post: 420826, member: 48050 wrote: I quickly quoted a couple places today, and the tooling for an order was $400-$500.

To me, that figure doesn't really seem to be that high; especially if you are considering making several. If you're only going to make one or two, then yes, the tooling cost would be significant. But, the more of these models you would make, the less that tooling figure would work out to be per unit, right?

Besides the tooling cost, which I'm assuming you would only pay for once, what would the actual cost be for each unit they fabricate?

DonnyThompson, post: 420835, member: 46114 wrote: I find fault with that statement. ;)

Don't let yourself think that. First of all, it may not be true. Have confidence in yourself and accept the possibility that the ideas you've come up with could be very good ones. Second, consider the possibility that "no one has done it yet" does not mean that it's not do-able.

It may in fact be very difficult. But that shouldn't stop you, either.

If you really want to do this, you should.

Oh...and one more thing....You're in mighty good hands with Boswell. To say that he is highly intelligent would be an understatement. ;)

Ok. Fine. I'll do it. Sheesh. ;) I think I have a plan for how to make the switching work.

Also, yes, Boswell is a guru. I very much appreciate his and everybody's advice.

DonnyThompson, post: 420841, member: 46114 wrote: To me, that figure doesn't really seem to be that high; especially if you are considering making several. If you're only going to make one or two, then yes, the tooling cost would be significant. But, the more of these models you would make, the less that tooling figure would work out to be per unit, right?

Besides the tooling cost, which I'm assuming you would only pay for once, what would the actual cost be for each unit they fabricate?

The PCBs are relatively inexpensive, ~$20-$30 once the tooling is paid. You're right, it's definitely more cost effective the more you get. I work with a local board shop at my day job, so I could probably get them to hang onto the tooling. I haven't designed a PCB in a few years, so that could be fun again, too.

Then I would just need a custom chassis. No idea what that's going to cost.

I'm lucky my other half is supportive. "You want to build a what now?...What's that?... Ok, sounds like a great idea."

pcrecord, post: 420837, member: 46460 wrote: I've go some grand-child version of the [="http://www.uaudio.com/hardware/compressors/1176ln.html"]1176[/]="http://www.uaudio.c…"]1176[/] in my [[url=http://="http://www.uaudio.c…"]UA[/]="http://www.uaudio.c…"]UA[/] [[url=http://[/URL]="http://www.uaudio.c…"]4-710[/]="http://www.uaudio.c…"]4-710[/]. Having a, stereo, closer to the original version would be fantastic (if affordable)

Affordable is the trick. I'd probably be looking in the $600-$700 BOM cost per channel. I know the 4-710 has limited versions of the 1176, but it's less than that per channel, with the pres! UA's dual 1176 (Rev D) goes for $2800. Funnily enough, one of the reviews of it on UA's site suggests basically this idea!

SSL

The cart is probably before the horse at this point. It's time to get out the catalog, dust off the soldering iron, and get to it.
Thanks for the input all, I'll post updates of progress.

I've read a few reviews, it's funny that some say that the new version don't sound like the old ones. Thing is, put 12 old ones in a row and none of them will sound the same. I bet new will sound like an old one in 30years... who knows.

I never buy gear because they sound like anything. I just want it to sound good.
If your units are any near musical as the newly issued ones, you'll get many sells ;)

Back to the 1176 Rev L ...

I've done some more work on comparing the A and D schematics and think that a switched version would be possible. It makes best operational sense to use the same gain-reduction (GR) and metering FETs in both modes, so there is some additional complication due to switching the FET connections around.

If you used a multi-way push switch (I think 8-pole DT would be enough) mounted on the 1176 front or rear panel, this could support the new small board. The type of switch used in mixing consoles would be suitable, and an example is the [[url=http://[/URL]="http://www.ebay.com…"]TA8UEE[/]="http://www.ebay.com…"]TA8UEE[/]. Locating it through a hole drilled in the rear panel would be better visually, but may result in excessive lead lengths internal to the 1176 chassis. Customers would have to be advised that changeover between A and D modes of operation must be done with the power off.

By my reckoning, it would need a 26-way ribbon cable connector on the new board with the mating ribbon cable split out to flying wires connecting to the D board below. On this D board, there would need to be something like 8 track cuts to match the 8 poles of the switch. If building the D board from scratch, many of the track cuts could be avoided by modified component insertion. The GR and metering FETs would move to the new board.

The metering circuit differences are a pain, mainly because you would want to keep two sets of calibration adjustments distinct for the two modes of operation to avoid the need for re-calibration every time you switched over. It means that most of the metering circuit would have to be duplicated on the new board.

If the stereo link components were to be added to the new board, there would need to be something like a 3.5mm jack on the rear panel to receive a linking lead that connects the two chasses together.

I won't do any more on this for now, but let me know how enthusiastic you feel about tackling the project (if at all) once the implied amount of work has sunk in!

very cool, Bos!

I ordered a preliminary BOM on Tuesday, so I haven't been frightened off yet. Jumping the gun a bit, but if I waited until all the kinks were worked out it probably would never happen. I live very close to Digi-Key's warehouse, which stocks an enormous amount of components, so more parts can always be obtained quickly.

I decided to order both Rev A and Rev D PCBs, and am going to utilize what I need of the Rev A board.

I haven't ordered a switch yet, but was looking at something like the TA8UEE. I think the calibration may be the death of me, as the number of switches required jumps rapidly if I start switching out pots. The bias pot looks like it can be left alone, but there are three pots in the metering circuit! It wouldn't be the end of the world to have to use two switches for a prototype model, but I'm looking for a more elegant solution to that issue.

Requiring power off to switch revs is somewhat problematic, but I don't have a way around that for now... another one think to think on.

I'm having fun, anyway.

Bos, thanks again for your valuable insight.

The 8-pole switch included enough poles to switch the metering circuit as well as the signal circuit. However, you can't tell from the way the schematics are drawn which meter pots are pre-sets on the PCB and which are user-accessible panel adjustments, so this is not certain.

When you said you had bought both the A and D rev boards, was this just the fibreglass or did it include the components as well?

I mentioned switching off the power when changing mode principally to avoid large transients in the output rather than any inherent need within the 1176 itself.
It would not be impossible to have something like a momentary push switch that clamped the output next to the mode change switch, and the user should keep that depressed when switching mode with the power on.

When you said you had bought both the A and D rev boards, was this just the fibreglass or did it include the components as well?
[/quote wrote:

I bought the bare boards for both revisions, components for all of Rev D, and the components I need from the Rev A build. Plus extra transistors for gain matching.

[QUOTE=
I mentioned switching off the power when changing mode principally to avoid large transients in the output rather than any inherent need within the 1176 itself.
It would not be impossible to have something like a momentary push switch that clamped the output next to the mode change switch, and the user should keep that depressed when switching mode with the power on.

I like it. Or possibly some sort of time-constant based clamp as part of the main switch. A two-stage switch would be great if one were available with this many poles. I've heard some people don't like large transients flowing through expensive audio gear. I'd hate to be responsible for someone not following directions and blowing up something downstream.[/QUOTE][/QUOTE]

Reverend Lucas, post: 421031, member: 48050 wrote: I ordered a preliminary BOM on Tuesday, so I haven't been frightened off yet. Jumping the gun a bit, but if I waited until all the kinks were worked out it probably would never happen. I live very close to Digi-Key's warehouse, which stocks an enormous amount of components, so more parts can always be obtained quickly.

I decided to order both Rev A and Rev D PCBs, and am going to utilize what I need of the Rev A board.

I haven't ordered a switch yet, but was looking at something like the TA8UEE. I think the calibration may be the death of me, as the number of switches required jumps rapidly if I start switching out pots. The bias pot looks like it can be left alone, but there are three pots in the metering circuit! It wouldn't be the end of the world to have to use two switches for a prototype model, but I'm looking for a more elegant solution to that issue.

Requiring power off to switch revs is somewhat problematic, but I don't have a way around that for now... another one think to think on.

I'm having fun, anyway.

Bos, thanks again for your valuable insight.

It's okay... be patient and do it right. I can wait to get mine. ;)

You mean Universal Audio offers 1176 kits? Are they a lot cheaper that way?

UA doesn't offer the kits, but the original circuit was designed in the 60s, so its patent is expired. Hairball Audio offers a replica of the original designs, with modernized power supplies.