Thomas R. Dial
The Soundcraft Spirit Studio recording console was originally released in 1990. According to my research, two frame sizes were provided: a 16 channel frame and a 24 channel frame. Targeting the home studio market, the Spirit Studio was an "inline" console where each channel has dual inputs: mic/line, and tape return.
The console also features direct outputs on each channel, 8 group buses, 4-band EQ including two parametric bands (and the EQ's can be split between mic/line and tape returns as well), and a host of effects sends as well as two "foldback" monitor mix sends. All in all, the feature list is quite impressive!
A console of this type was likely to be paired with a semi-pro 8-track recorder such as those available from Tascam. but around the same time, Alesis released the ADAT, which put digital multitrack recording in the hands of the masses for the first time. Since ADATs could be linked together with sample-accurate synchronization, 16 and even 24 track studios came into reach for serious home musicians and project studios, and along with completely hard-disk / DAW based systems, were probably the death knell for some small professional studios (citation needed.)
For years, I've operated without a hardware mixer, opting instead to expand my I/O capabilities and record synths directly into Pro Tools and Logic. However, when I became interested in the craft of mixing, I noted that many of the engineers I admired had "grown up" in the era of analog consoles and had developed plenty of intuition in terms of gain staging and EQ by using analog consoles in the past (even when many of them have since moved on.)
So, because I love electronics projects, I decided to buy a reasonably decent analog mixer cheaply and restore it. Furthermore, I decided to document the restoration project so that interested parties may learn from my experience and potentially, avoid my mistakes!
If you're planning on a console restoration project, I already have a few suggestions for things that you should consider.
-
Join forums These days, there are plenty of online forums for people with similar interests. Forums can be a good place to learn about the pros and cons of various pieces of gear. I say "can" because there are also a significant number of people who speak with authority who actually have no idea what they are talking about, so be cautious. For me, the key value of forums is gaining insight into potential pitfalls, common issues, and potentially, sources of spare parts.
-
Shop locally Recording consoles like this are large and heavy. Even the sixteen channel console that I am restoring is large, and I ended up paying $100 just in shipping. This was greatly offset by the low price that I paid, but I'd encourage you to shop locally where possible for two reasons. First, shipping costs. Second, you'll have the opportunity to test the console out and see its condition before purchasing and incurring shipping costs. In my case, I really lucked out as the console functioned very well (nothing broken, at least) and wasn't missing any knobs, sliders, etc.
-
Make sure you have schematics Schematics can be hard to find but are necessary if you truly need to dig in with the restoration. They are also very interesting to read and may give you good ideas if you're interested in building projects of your own design.
-
Consider parts availability Parts for some consoles can be hard to locate. A friend of mine has a Calrec console that uses ICs that are no longer being made. Obtaining spares will be expensive. By contrast, the Spirit Studio uses (mostly) still-available op-amps.
-
Ensure the power supply is reliable Semi-pro and professional consoles will typically have external power supplies. If you're working on an older console, you will want to make sure that the power supply is in good shape before powering on. While a quality power supply will provide adequate protection for faults, it would be bad news if, for example, one of the power rails in a bipolar supply failed and the other one remained functional. Check large electrolytic capacitors (more on that later) and replace if bulging, leaking, etc.
NOTE: I've seen some consoles for sale without their power supplies. I would be wary of this. If the seller doesn't have the power supply, there is no real way for you to test it. Additionally, these power supplies are speciality items and often sell on eBay for hundreds of dollars. I got my console (and power supply) for less than what other people are selling their power supplies for on eBay!
-
Have the correct tools, know how to use them Restoration of a console will generally require some parts to be repaired or replaced. This, in turn, requires desoldering and soldering. You need to have the right tools as well the expertise for this. If you don't, you have two options: pay a qualified tech, or teach yourself. If you opt for the latter, practice using the tools and techniques on old circuit boards before touching any gear you care about. I would strongly encourage someone undertaking a restoration to spend the time learning how to do these things yourself. Otherwise, whatever you think it will cost to restore the console will be multiplied by some factor.
-
Establish a budget I alluded to this above, but restoration of a console can and will get costly. For example, I am considering whether to "re-cap" the Spirit Studio. If I do, it's going to involve replacing in the area of, say, 500 capacitors give or take. Not only will this get expensive, it may actually be hard to source the caps given the supply-chain shortages in 2021. If you're new to analog consoles, the budget will also include cabling to integrate it into your system. You may also need to upgrade your computer interface (assuming you use a DAW-based system) to add channels in support of integrating the analog console.
0.5 Hours
The console arrived today, and I had some help unboxing it from my Corgi pal, Stanley, who does not look like he approves of the purchase in this picture. Perhaps he thinks it will mean less time spent with him?
Despite the fact that I knew the dimensions of the console ahead of time, I was rather surprised at its size. I was also surprised by what seems like extremely robust, thick-gauge steel construction. While I've seen some considerable shade thrown at the Spirit Studio in forums, often times I feel like people are conflating them with later plastic models that are targeted at more budget studios.
Case in point, This article from June 1991 lists the price of the 16 channel console as 1651 GBP, which in 1991, according to some spot currency rates from that time, translated to $2734 USD, which adjusted for inflation translates to approximately $5257 in 2021 dollars!
So, this was not a casual purchase for anyone in 1991, and was not strictly a budget console either. In fact, it is on par, price-wise, with a Toft Audio Designs ATB16, which is probably the closest thing that you can find to the Spirit Studio 16 today.
Here it is out of the box (above.)
4.0 Hours
As noted, I am behind on documenting progress with the restoration project, but am working through the backlog. The entry for this day documents going through the console's features to see what, if any, trouble spots exist.
And this console has a lot of things to test. A total of 16 input modules, each with a mic/line preamp, tape return, insert, and direct out. Also a routing matrix to the group buses, 4-band equalization, dual input/tape signal paths, aux sends...The list goes on.
I am a huge fan of checklists, which I find incredibly useful in managing software projects at my day job as well as just my overall day-to-day. So, I created a task list with the goal of checking every single function of the console prior to disassembly. With so many things going on, it would really be impossible to keep it all in my head.
Here's what it looked like:
This picture was taken before I was done. I made an attempt to capture every single feature of the board (the picture merely depicts some of the features for an input module) and then went through, channel by channel, feature by feature, and marked the sheet accordingly.
I was pretty astonished by the end to find that the console has no faults except for the occasional scratchy switch and/or dirty potentionmeter. All the connectors, switches, pots, faders, LEDs and features work as expected. It's almost hard to believe, given that this console is 30 years old, but I'll take all the good news I can get!
One thing I noticed while inspecting further is that the console has been dropped at some point, perhaps even during its most recent shipment. However, Nothing internally seems to have been damaged, and I am pretty sure that I will be able to bend the heavy gauge metal back into shape with some stratgic use of clamps and blocks of wood, etc.
5.0 Hours
After using my new ESR meter to test some capacitors, I came to the conclusion
that they are out of spec. While it is impractical to test all of them, I
pulled some out of circuit by desoldering them in order to ensure that the
test was as accurate as possible.
Generally, electrolytic capacitors get old after awhile. And my Spirit Studio is about 30 years old at this point. It's time to bite the bullet.
Replacing capacitors is a job that should not be undertaken lightly. You really need to make a serious commitment of both time and money. In my case, I did a ballpark estimate and figured it would cost me around $200 to replace all of the capacitors on the board.
Next step: actually count how many of which value I need and put the order in.
As shown in the photo, I proceeded by pulling one of each type of card from the console, and then made checklists for all of the boards. Shown in the photo are a group PCB and the master section PCB.
Here's an example of a checklist:
The checklist includes the PCB part label (for example, C21 on the PCB) and the value in microfarads as well as the voltage. I made lists for each of the types of PCB in the console:
- Input Module (x16)
- Group Module (x4)
- Group Module Meter Daughter Card (x4)
- Master Module (x1)
- Master Module Meter Daugther Card (x1)
Once I completed (and double checked!) all of these lists, I transferred the data to a spreadsheet. In the spreadsheet, I sorted all of the items by part and used formulas to calculate how many of each part I needed to order from my supplier (Mouser.)
Here's the sheet:
Spirit Studio Electrolytic Capacitors
Note that the spreadsheet is in the ODS format. That's because I use Linux and LibreOffice (as opposed to Excel for Windows or Numbers of Mac). If anyone is interested in the sheet in Excel format, let me know and I'll export it.
At the bottom of the sheet are the totals. When it came to order, I actually ordered about 20% or more extras. Why? Capacitors are in short supply right now with the 2021 pandemmic-related supply-chain shortages. Also, it's possible that some will be defective and so I wanted to order some extras just in case.
When you're ordering in these quantities, the per-unit prices fall pretty quickly.
The total tally for my capacitor purchase was around $180 USD. This included some that I was forced to source on eBay because they were out of stock at Mouser and wouldn't be delivered until late November. I am not interested in waiting that long, so I put in the $10 eBay order.
2.0 Hours
This is a sort of phony update since it was actually one of the first chores I took on: disassembly. By disassembly, I really mean:
- Removing all the knobs / switch caps / fader caps
- Basic external cleaning
- Removal of one channel card to check the state of a fader
This console had clearly been used. When it arrived, it had a fair amount of dust and dirt around the potentiometers and faders. The knob, fader, and switch caps were also fairly dirty. The faders felt OK, but a little dodgy and possibly dirty as well. I wanted to take a look.
I started by removing every single knob, fader, and switch cap, I put them all into a plastic bin with the intent to soak them in soapy water. However, before doing that, I tested a single knob with Palmolive hand-dish detergent as well as powdered machine detergent. I have read about unfortunate times when people attempting to clean old plastics end up ruining them unwittingly with chemicals, so I wanted to ensure the soap I used was compatible.
Luckily, after soaking for a night in each soap, the test knobs were fine: spotless in fact. Because the machine detergent seems easier to rinse, I proceeded with that and filled the entire bin with some machine detergent and warm water, and let it soak.
Next, I opened the case. It's held together with metal screws. On my console, the plastic side pieces (which are normally screwed to the console via threaded brass inserts in the plastic) were broken, so all I had to do was remove screws from the case.
Once opened, I removed the hex washers for all jacks and potentiometers for the first channel card. There are two screws holding the channel's fader in, and another screw that must be removed. Following that, you must remove a twisted-pair cable from the channel card. Finally, you must remove the ribbon cable connecting the card to the bus (Note: I'll add pictures of this soon.)
TIP: Make sure to take pictures as you go along, then you will never forget how things were held together.
I mentioned earlier that I wanted to check the state of a fader. A key reason for this is that if they are totally shot, I need to start sourcing replacements right away.
The channel faders are attached to the channel card with three wires, as shown in the picture above. After noting their position I elected to desolder the wires to make handling the fader easier.
Below is a fader after being disconnected. I noted that it was indeed dirty, and the knob didn't move smoothly in the track. It felt as though there was dirt in the track. Linear potentiometers are always prone to dirt and dust, and this was no exception: I could literally pull dust bunnies from inside the fader!
The next step was to disassemble the fader itself. To do this you must work very carefully. I used a pair of needle-nose pliers to carefully bend back the tabs holding the fader assembly together. (These tabs should be visible in the photo.)
Once disassembled, I cleaned the parts using isopropyl alcohol and a couple Q-Tips. The track on the fader appeared to be in O.K. condition. I was careful to note the positions of all the pieces before taking faders apart: the digital camera is your friend here!
Once clean, I lubricated the carbon track on the fader PCB with Deoxit F5 Faderlube. I also applied some Deoxit fader grease to the inside of the metal case which the wiper slides in. Just a little grease is enough.
After reassembly, the fader felt as good as new and definitely better than the others. I am hoping that the others are simply dirty and can be restored like this.
I ended up reassembling the console after all this (although the knobs and switch caps are still in the plastic bin, now dry.) In a later step, I will go over cleaning / refurbishing of the case itself, and while I am waiting for capacitors to arrive, I'll start thinking about how I want to replace the side panels (probably with some hardwood.)
0.5 Hours
Audio consoles in this price range are not fully modular in the sense that a true professional console is. The Soundcraft Spirit Studio (and even its successor, the Ghost) employed a monolithic chassis to save costs, but retained individual module PCBs. In a modular console, my understanding is that there is a "back plane" type of arrangement so that modules have edge connectors that plug into the chassis. These connectors are probably fairly robust and also expensive.
Again, to safe costs and space, consoles in the "home studio" price range use low-cost edge connectors and connect the modules with a ribbon cable. These cables are often the source of failures and intermittent connections when they get old, but can also be difficult to replace. While mine is in good shape and all of the functions checked out on my console, I have elected to do some basic cleaning to make sure connections are good.
The picture above shows what the ribbon cable looked like when it was removed from the console. As you may be able to see, it is dusty but there also seems to be some kind of white residue on the connectors in some areas. I am not sure what this is, but I decided to clean it with isopropyl alcohol (IPA). I used the purest form I could buy at the local drug store (91%) and an old tooth brush.
Cleaning with IPA had good results, as you can see above. The cable looks as good as new. I plan to also apply Deoxit to the edge connector pins and also to the cable connectors prior to re-assembly.
One last note: As discussed, it is apparently difficult to find these cables, and I am told that making reliable replacements at home might prove difficult. For this reason, I am aiming to minimize my handling of the cable. I will assemble the console fully to test the first channel card that I repair during my dry-run of that process, but I won't test each card that way.
My hope is that I can avoid connecting and disconnecting the ribbon as much as possible as not to wear out the connections.
Some follow-up tasks:
- Contact Creation Audio Labs to see if they can create new ribbon cables
- Build a powered test-rig for channel cards so that I can test the cards
- Inventory - 0.5 Hours
- Planning - 0.5 Hours
- Implementing ESR Calculator - 0.25 Hours
This past Thursday, my capacitor order arrived. As discussed in a previous section, in order to replace the capacitors in the console, I had to order them in significant quantity. I elected to choose Nichicon UFG ("Fine Gold Muse") capacitors, which are nice audio capacitors and quite suitable for replacing the old capacitors in this console.
I ordered approximately 20% more capacitors than I needed in order to allow for out-of-spec parts, damage, and errors in delivery. I can use the spares in other projects or possibly just sell them on eBay. Purchasing in bulk, going over on the purchase quantity didn't cost that much and definitely added to peace-of-mind in case I counted incorrectly.
My first step was to take inventory, and I found that Mouser is extremely accurate in terms of shipping. I had one extra capacitor in one value and was missing another in another value. However, this amounts to a very small error and my over-order corrected for the missing capacitor.
At this point, things are "getting real" with the project. I have done a fair amount of pre-work to label the cards, I've removed all knobs and fader caps, and have assessed what I will attempt to address and what I will not.
Since I'm about to embark on a major effort to replace all of the electrolytic capacitors in the board, I need to perform a test to ensure that nothing will go horribly wrong. So, my approach will be to service a single channel card first, reassemble the console, and test the channel. If all goes as well with the first channel, then I will proceed.
I have therefor removed the first channel card and will proceed with making a checklist of items to address, and then will go ahead with that. So, what have I decided to do, and what have I decided not to do for each channel?
-
Tasks
- Jacks - Will be removed and ultrasonically cleaned in IPA.
- Switches - Will be removed and ultransonically cleaned in IPA.
- Faders - Will be overhauled and replaced if necessary.
- Edge Connector - Will be cleaned to ensure good contact with ribbon cable
- Electrolytic Capacitors - Will be replaced
- Potentiometers - Will be cleaned externally
- Miscellaneous - Spot repairs made as needed (some ICs are not seated well.)
-
Non-tasks
- Potentiometers are in good shape; no rebuild or overhaul will be done.
While it seems like extreme overkill, I am trying to document this project in detail as an exercise in project management, so I will be creating a detailed repair checklist that will be included in the resources section.
A final note: I wrote a small C program to calculate expected ESR at a given test frequency. This program relies on values provided on the capacitor's data sheet. I will cover this program in a later post.
Now that I have all of the tools, parts and resources ready to work on the console, I've elected to conduct a trial run to see how servicing a single channel goes. My plan was to pull the first channel card, perform all the tasks mentioned in the previous post, reintall it, and test it.
After pulling the card, I started by depopulating the components from the PCB that I planned to drop into the ultrasonic cleaner. I found a scrap of anti-static foam that I'd purchased for storing sensitive ICs and decided to use it to hold parts that had been removed.
All of the switches and jacks were desoldered during this step. Desoldering was done with a Hakko FR-301 handheld desoldering pump. At this time, it's worth saying: don't attempt a project like this without an automatic desoldering pump, and make sure it's a high-quality unit.
I've spoken to experienced techs and friends who work in the industry, and several of them have told me how efficiently they (or people they know) can work with desoldering braid or using other techniques.
Let's just say this. I'm skeptical.
Short of posting a video (which I may do if I get around to it) I just don't think there's any way in the world that a person working with braid or any other manual method could beat someone skilled with a desoldering pump. With braid, spring-loaded pumps (which you should definitely avoid!) or any other method, a job like this is going to take you much longer.
Furthermore, don't be cheap when buying tools. Prior to the Hakko, I had an Aoyue desoldering station that I purchased for less than $200. It's sort of a clone of the Hakko in terms of its overall design, but has a very large metal desktop case for the pump. The craftmanship of the Aoyue was quite poor indeed; the gun felt like it was going to fall apart, and things that were supposed to have a good seal did not. Additionally, it seemed to lose suction rapidly.
With the Hakko, I serviced an entire board (removing all switches, all 18 electrolytics, and all jacks) and everything worked perfectly, every time. I didn't lift a single pad, and parts practically fell out on their own.
TL;DR - Buy a Hakko FR-301, at a minimum.
Once all the parts were removed, I dropped the switches and jacks into an ultrasonic cleaner with 99% isopropyl alcohol, purchased online. The ultrasonic cleaner does wonders with the parts, and they appeared new after cleaning. Using a high-grade alcohol is recommended for cleaning electronics / PCBs, but make sure it's compatible with plastics you're cleaning first.
The picture above depicts the parts after cleaning.
Since I had already handled overhauling the fader for this card (I wanted to assess them in order to see if I needed to order spares) the next step was replacing capacitors. Here, I worked from the checklist and replaced caps more or less one by one.
I went very slowly so that I could check the specs of the pulled caps. Was it necessary to replace all of the capacitors? Well, almost all of those that I pulled were in the correct value range, but most had significantly higher ESR than the replacements. My conclusion: since the board left the factory 30 years ago and was obviously used a lot, and since the capacitors showed higher ESR than they should, replacing the caps with top-notch Nichicons was a good call and will buy me years of service.
The picture above depicts the completed channel card, with the switches reinstalled as well as the Nichicon gold capacitors. As a final step, I cleaned the mask side of the PCB with IPA to clean up residue that accumulated during soldering. I also visually inspected all of the joints to ensure that they were complete.
While it's painful to keep removing and installing things from the console, my plan was always to put one of the channels to the test before proceeding. Now that I've completed the overhaul of one channel, it's time to proceed.
After reinstalling the card, I connected the bus lead and the ribbon cable, which I had completely removed previously for inspection and cleaning.
After powering on and letting it warm up, I plugged in my Akai MPC-2500 sampler and loaded a drum kit. I wired the sampler's stereo outs into the first and second channel's line inputs, and routed the channels to a bus. After a few mishaps with switches not being depressed properly (it's hard to tell what position they are in sometimes, especially without the switch caps) I proceeded to run through the features.
I didn't expect to be able to hear a difference, but one thing that was quite obvious was that the bus switches, which were noisy before, were whisper quiet when activating and deactivating. Apparently the ultransonic cleaning worked! I also noted that the fader had a much smoother feel than the rest (which were not overhauled.)
After running through tests of the EQ, and testing all jacks, I am quite satisfied that the overhaul of the channel was a complete success! Next steps will be:
- Pull remaining cards and prepare for servicing them all
- Clean the case, removing grime, adhesive, stickers, etc.
- Hammer out a bend in the case (near the left side.)
Revisiting some earlier pictures of the console in this document, you can observe that this was a well-loved piece of gear. Not only was it dirty from years of use, the previous owner had used masking tape over the the scribble strips to mark channels, affixed stickers, and so on. It also seems that the console was dropped hard on the left side at one point, seriously denting the front piece and bending the case enough to be noticable when using the channel one fader.
My belief was that using an array of household products and tools, I could get the case back to shape. Here's a list of the tools/products used:
- Goo Gone (TM)
- Acetone (nail polish remover)
- Water
- Razor Blade
- Piece of hardwood
- Hammer
- Magic Eraser (TM)
My first step was to test the chemicals in an unobtrusive area (the back of the case) to make sure they wouldn't remove the paint. No probem was observed. Next I moved on to try the chemicals on the "Spirit Studio" emblem on the right-hand side of the console. My logic here was that if either Goo Gone or acetone would remove the paint, I'd be better off removing the emblem than legends or markings on channel strips.
That test went well also.
The rest was all elbow grease. I pulled off the tape by hand, using the razor (carefully, but not carefully enough, for I did accidentally scratch the paint on the left hand side: more on that later.)
After the tape and stickers were removed, I went after the affected areas with Goo Gone. It works great! So great in fact, that I decided to wipe down the entire console surface with it. Years of grime and other little smudges dissolved effortlessly. I didn't overdo it, though, and spent much more time working on the areas covered with adhesive.
Finally, it appears as though the previous owner used Sharpie to mark channel numbers directly on some of the scribble strips. These marks didn't come off with Goo Gone, so I went after them with the acetone nail polish remover. Voila! The acetone removed the marker but left the paint.
Finally, using spare blocks of wood and a hammer, I hammered the bent metal back into shape. I neglected to take before and after pictures here, but it is probably visible from the picture above that there are no major bends left. I am quite satisfied with the results and ready to proceed with the rest of the overhaul!
I don't have much in the way of updates lately. Work and the holiday season have slowed me down to an extent, so I am just slogging through on performing the maintenance of each of the channels. So far, however, I am about halfway done with the project; I've completed twelve input modules so far.
One thing worth mentioning, if you're embarking on a project like this and are also using an automatic desoldering pump, is to follow the cleaning procedure for the pump early and often to ensure that it is functioning well. Otherwise, you could face loss of suction.
The Hakko FR-301 comes with the tools necessary to clean the unit. I recommend cleaning it every time you're done, before you put it away. I neglected to do this across two or three evenings while I was working on a bunch of input channels, and the suction got noticeably worse before I remembered that I hadn't cleaned it in awhile.
Something else that I forgot with the project was that I planned to perform the user modifications to convert the console's tape sends to +4dBu (they are -10dBu by default). Luckily, I think the resistors that must be snipped to accomplish this are in a position where I can do it without pulling the cards. I hope so!
Until next time, happy holidays! Soon, I'll be done with input channels and will be ready to move onto the group cards and master section. Then, it will be time for the system test!