Help finding schematics for an old PCB

[jacobtohahn]

Hey everyone,
About a year ago, I picked up an 8" floppy disk drive on eBay, with an exclosure. The enclosure takes in wall power, puts it through a transformer, and rectifies the resulting AC into two DC voltages using a PCB. The recification PCB is what I'm interested in. I don't know exact dates, but the board had to have been made from the late 70s to the early 80s. It seems inefficient for what it does, and I could very cheaply create a new one with modern components. However, I would first like to replicate the original design. The issue is that, when desoldering components to replace (the output voltages weren't stable at all), some pads lifted from the board very easily. Also, the legs of a TO-220 component were very fragile and the whole thing ended up falling off. Obviously, the board isn't in the best condition.
So, to help me design a new board and create a better replacement, I could use a schematic. After hours of searching, I haven't been able to find one anywhere (or even a single mention of the board) on the internet. Hopefully, someone here might know enough to help me find something.
Both the board and enclosure are made by RNB Enterprises Inc., which, I found out after some research, went out of business a few years ago.
My hope is to be able to find at least a board schematic, but if possible a drawing of the copper and board outline would be incredibly useful for replicating the design.

Here's everything I know so far:

• The board outputs two voltages. The board is split into two identical sections, one for each rail. I'm not quite sure what the levels were supposed to be, but one rail goes to the logic board and one to the drive head motor.
• There is an AC input for each rail and a DC output for each.
• Besides passives, it uses two full bridge rectifiers, two LM350T regulators (which most of the passives are for), and a TIC116F thyristor.
• The two large capacitors were not present with the board. However, the seller had one which had been removed, so I bought exact replacements from eBay.
• Top board text: "OBC 94V-0", "RNB ENTERPRISES", "POWER SUPPLY 2000"
• Bottom board text: "RNB ENT", "PWB 2000"

Here are some photos. I have info in the caption of each photo. It should be helpful and is more specific than the list above:
https://imgur.com/a/Ccez51a

 

[ldkraemer]

Some pads lifted from the board very easily!

Yes, that is from to much heat for too long a time. A temperature controlled iron plus more
experience will lessen those type accidents.

Let's see if you can get the Pad(s) repaired first and go from there. For the pads that you have lifted
grab your drill index and carefully take smaller bits and keep slipping the bits through the hole until
you find one that just matches the size of the hole. Don't enlarge the hole or use a motorized device
to get the drill bit through the hole. Now measure the size if the drill bit with your Caliper or Micrometer.
Post the exact size of the drill bit(s). We will go from there.

Each power supply doesn't look too complex. You should be able to draw out the schematic, but it will
take a bit of time.

Larry

 

[jacobtohahn]

Yes, that is from to much heat for too long a time. A temperature controlled iron plus more
experience will lessen those type accidents.

Of course. I have a soldering station and I made sure to keep it at low heat. My theory is that either the adhesive bonding the copper either wasn’t the best from the start or that it broke down some over time.

Let's see if you can get the Pad(s) repaired first and go from there.

Well, my goal here isn’t necessarily to repair the board. If you look closely at the images, you can see that when the pads lifted, an entire trace connecting them fell off. Yikes. I’ve done tests with modern professional PCBs that I had designed, and even 6 mil traces came off less easily.

Tracing the schematic shouldn’t be hard, as you had said, especially due to the simplicity of the circuit. What I would like to do most at this point is to have two options, being to replicate the original board (for historical accuracy) or to make a more efficient and smaller replacement.

 

[Chuck(G)]

Just a thought tossed out...

If your purpose is to provide power to a couple of 8" drives, I'd be more inclined to bypass the old linear supply hassle and use an inexpensive SMPSU, such as this one. Your box will run cooler, too.

But I've run afoul of the "I want to keep it original, warts and all" crowd, so take this as a suggestion only.

 

[maxtherabbit]

That mean well is exactly what I'm using for my 8" setup, works great. I've put it on the scope as well and it has very, very low ripple

 

[jacobtohahn]

Just a thought tossed out...

If your purpose is to provide power to a couple of 8" drives, I'd be more inclined to bypass the old linear supply hassle and use an inexpensive SMPSU, such as this one. Your box will run cooler, too.

But I've run afoul of the "I want to keep it original, warts and all" crowd, so take this as a suggestion only.

That's a good suggestion. I would rather keep the original board (especially because my drive enclosure is missing the cover) and the transformer as well (I plan clean it up and remove the rust). Once I work out how to create a new cover, it wouldn't be as much of an issue. However, as it would be difficult to repair the board I have now, I would likely design a new one with the same dimensions. Is it that bad to use a linear regulator here, especially because I'm only powering one drive with the enclosure as a heat sink?
My ideal scenario is to be able to have copper drawings to accurately replicate the original board. I know this isn't very feasible, so I'm open to other options as well. If all else fails, just being able to get power for testing is the base goal, and that Mean Well would work good for that.

 

[Chuck(G)]

One advantage of the switcher is the higher efficiency and lower heat dissipation. There should be plenty of reference designs on the web for a linear supply, as they're older than the hills. Check into some of the bitsavers docs on, say IMSAI. Compupro and Altair. This ain't rocket science.

 

[jacobtohahn]

One advantage of the switcher is the higher efficiency and lower heat dissipation. There should be plenty of reference designs on the web for a linear supply, as they're older than the hills. Check into some of the bitsavers docs on, say IMSAI. Compupro and Altair. This ain't rocket science.

Absolutely, but here I don't truly need the greatest efficiency and the lowest power dissipation. This drive won't be running incredibly long. I know quite a bit about linear supplies and their workings, and I've used them quite a bit. I'm not the most comfortable with switching supplies, but it's good that most are already put together for you. I drew out the schematic for the original board, and it's exactly as I had expected. The regulators are connected with pretty much the exact schematic that is provided in the datasheet. The diodes are for protection, but from what I could find, they aren't absolutely necessary in this circuit. Better safe than sorry. It also looks like one regulator (probably the lower voltage side? Or maybe the higher voltage, as it would be going to just a motor which is less sensitive than a logic board) is lacking the decoupling cap for the ADJ pin of the regulator, which is again not the most necessary, along with the diode used to protect against discharge from that cap. The thyristor is used in a crowbar overvoltage protection circuit, which essentially just shorts the contacts and blows the fuse if something's not right. This happened once when I was first testing it. I may just use the schematic I drew and layout a new board with the same dimensions myself. Does that sound like a good option to you?

 

[Chuck(G)]

Sure--whatever works. At least one of my 8" drives is powered by the guts from an old open-frame linear supply--uses an LM723 to drive a large TO-3 NPN for the +24 with some overcurrent protection. Complete overkill, but the PSU was a surplus special and so cost almost nothing. My recollection is that the IMSAI 8" supplies weren't much more than a couple of bridge rectifiers, smoothing caps and linear regulators. Really basic stuff. Just watch the heat sinking on the +24 regulator.

 

[jacobtohahn]

Sure--whatever works. At least one of my 8" drives is powered by the guts from an old open-frame linear supply--uses an LM723 to drive a large TO-3 NPN for the +24 with some overcurrent protection. Complete overkill, but the PSU was a surplus special and so cost almost nothing. My recollection is that the IMSAI 8" supplies weren't much more than a couple of bridge rectifiers, smoothing caps and linear regulators. Really basic stuff. Just watch the heat sinking on the +24 regulator.

I took a look at the datasheet for the LM723 and immediately thought the same. It does look like a powerful chip, although complicated, but I might keep it in mind for other projects where I need good regulation.
The heat sinking should be fine. The case itself is used as a heat sink, which is plenty of thermal mass.
With your drive, what computer are you hooking it up to? If I can't get this thing to interface with my Zenith Z100, I might need to find a different solution.

 

[Chuck(G)]

My drives are hooked to various setups, including PCs, microcontrollers, and specialized boards, such as the Catweasel. They need to be--it's what I do.

 

[DeltaDon]

I've run into a faulty linear supply on the S-100 system in my avatar that took out the 8" floppies too. My fault, I guess. I did a visual check after receiving the computer and didn't see the broken lead on the axial main filter capacitor (hidden behind some wiring) that was broken just above the PCB. I guess caused by shipping gorillas tossing the box once too often. Magic smoke was let out of the drives. It was easy to figure the circuit out and make a copy of the schematic. While it is now fixed, I like the idea of replacing it with a Meanwell since there's no over voltage or other protection built into the original design other than what the 723 IC provided and that's not as easy to add as it is to just swap with a modern switcher design. Easy to do in this computer since there's two complete power supplies - one for the drives and a second one to power the S-100 cage. Two transformers and all.

 

[jacobtohahn]

I've run into a faulty linear supply on the S-100 system in my avatar that took out the 8" floppies too. My fault, I guess. I did a visual check after receiving the computer and didn't see the broken lead on the axial main filter capacitor (hidden behind some wiring) that was broken just above the PCB. I guess caused by shipping gorillas tossing the box once too often. Magic smoke was let out of the drives. It was easy to figure the circuit out and make a copy of the schematic. While it is now fixed, I like the idea of replacing it with a Meanwell since there's no over voltage or other protection built into the original design other than what the 723 IC provided and that's not as easy to add as it is to just swap with a modern switcher design. Easy to do in this computer since there's two complete power supplies - one for the drives and a second one to power the S-100 cage. Two transformers and all.

It's surprising that there's two supplies. My issue with replacing the supply with one like a Meanwell is that I like the original look of the supply and transformer. I'm going to clean the rust off of the transformer with a brush and sandpaper, and I think it will look good. I could design a switching regulator board with the same dimensions, but I know it wouldn't have the greatest efficiency due to my lack of design knowledge and the size of the board. I definitely don't want to be drilling holes in the enclosure to fit a smaller board.

 

[Chuck(G)]

Two supplies was par for the course in the early S100 days. The IMSAI dual floppy drive had two PSUs, because you could get it with a single floppy with a considerable cost savings.