Tascam M-___ Story...

Is this the answer to my power supply needs? This Hammond power transformer?

IMG_7693.JPG


We will know soon. In the meantime, I can tell you it indeed fits perfect. It's the smaller one setting there in front of the larger stock main transformer.

IMG_7695.JPG
 
Got the mounting holes drilled and prepped for the add-on power transformer, and got its inputs connected to the mains power. It is convenient there were two vacant solder lands on the power switch PCB. The vacant lands were labeled as being for a capacitor. I think the role of that cap is fulfilled by multiple caps on the power supply PCB, so I used 'em to elegantly tie the new transformer in. The new transformer is in the foreground, dwarfed by the factory main transformer in the background. I'm holding the power switch PCB tipped back so you can see where the new power transformer input leads tie in:

IMG_7751.JPG


With the power switch PCB reinstalled, I'm ready to snip some wires and patch in the new add-on transformer outputs to the power supply PCB.

IMG_7752.JPG
 
Frustrated.

I should have tested it before I drilled holes cut wires etc.

You would expect a 50V transformer rated at 50V RMS output to be...what...50V RMS right?

Well, it's about 58.4V RMS which means the output of the DC supply is about +/-40V...it's supposed to be +/-35V and if the 50V transformer output what it was specified to output the output of the DC supply would be just about dead on +/-35V. I think +/-40V is too high.

:facepalm::spank::wtf::cursing::mad::(
 
Shit. So how do you deal with that? Curse them out and send it back?

Install something else to drop the voltage to what you need?
 
Working through some different options for either regulating the circuits, or dropping the voltage using a big-ass zener diode.
 
The reason the transformer is outputting higher than 50V is because (duh) it is rated for 50V at 1 amp current load. It is sitting idle in the power supply. There's no way the the output boost circuitry will draw anywhere near that, even at the maximum possible. So if the circuit won't tolerate the +/-40V amplitude something will need to be done, and likely a series zener diode in the power supply rails. That would be a simple and effective solution. BUT my friend thinks the circuit will handle the 40V, and sometime this coming week he is going to mount the same parts from his spares on a piece of breadboard and see what happens when powered at 40V. :eek: pretty cool. And if it DOES handle 40V, that will mean the buss, aux and stereo outputs on the M-__ will have essentially incomparable headroom.
 
GEEEEEEZ 6 months goes by quick.

I've been so inundated with work and personal stuff...this project just sorta went on the back burner waiting for some follow up from my friend who's been a big help on and off...and with being busy I've been pretty lax about following up with him...plus, in all honesty when I have had time to work on projects I haven't felt much motivation to do anything.

Anyway, I do have a couple updates, but the simplest one first.

Awhile back I finally recognized what was right in front me the whole time, that the Tascam M-50 has a PCB assembly in it identical to one in the M-__, and it answers a nearly decade-long question about the +/-35V power rails labeled in the M-__...

Anyway, being one who likes having spares on-hand, and seeing as spares are a relatively unobtanium commodity with the M-__ being one-of-a-kind, I've been watching a couple auctions for these M-50 PCB assemblies on eBay that match the M-__...been watching for months. I finally made an offer on them. $24 for the pair shipped. I dunno...silly I suppose. But I got them today and probably will never do anything with them, but I just like having them...I don't expect I will ever do anything with them, but I guess it satisfies something in my brain to have the parts because the M-__ was designed with the capability to have balanced group and aux outs...there are lands on the PGM group PCB for another set of these ultra-high headroom line amps, and punch-outs on the backplane for the XLR output jacks. Anyway...I got them today. They look like this:

IMG_9403.JPG

I'll post more updates later.
 
Well, that's 'some' advancement. :D

I saw something on the m50? Nicely built piece of gear. What motivated getting back onto that? A cash desire?

:D
 
So here is probably the most important update...it is in regards to finally resolving not only questions about one of the bipolar power supply rails, but also implementing modifications to the supply to achieve the power needed as determined by the resolution to questions about that set of rails.

So this is kind of a long post, but so much time has gone by and a refresher may be helpful for readers as well as for me down the road...this thread has been invaluable to me to be able to re-read and see where I left off, what's resolved (and how) and what is pending (and why). Anyway...

So last Fall I finally (for the first time since I've had this console) determined what kind of power all the power rails were supposed to be. There is screening on the module cards in the mixer indicating the following DC power feeds this console: +6V, +/-15V, +/-35V and +48V. From the beginning I erroneously thought the +/-35V rails must be an unregulated supply feeding circuits with local regulation...the traces for that power went to these transistors with their own heatsink in each module. I assumed those were r regulators, and the local regulation stepped it down to 18V because I'd never seen anything higher than that in a piece of Teac/Tascam gear...anything with +4dBu output was powered by 18V. So I took the +/-18V supply in the PS-520 supply I mated to the M-__, pulled the regulators (making it +-/26V unregulated) and went on my merry way thinking the power was being regulated to +-/18V locally in each module for the buss and aux out line amps. Silly. That doesn't make any sense why they would do that, and why the screening in the mixer would call for +/-35V since that's WAY more than what is needed for +/-18V regulators to do their job...I just thought I had it all figured out and was comfy with that because it fit what I knew. But, and this should be no surprise, it was bugging me...the whole +/-35V thing. That amplitude did not fit logically with an +/-18V regulated circuit...unnecessary...Teac doesn't do weird unnecessary stuff like that. So I couldn't let it go and back in 2014 started asking my guru friend some questions. At that point once I started providing him with some pictures of the "regulators" and some coarse sketching of parts of the circuit he pointed out the "regulators" were part of a line amp. So then the questions were "Well if its a line amp, is it really supposed to be powered by +/-35V power, is the power regulated or unregulated, and does it *have* to be +/-35V? Could it be less? What's the advantage to the 35V amplitude?" It was really confusing to me because the line amp circuit also includes +/-15V powered opamps...it was a really confusing thing to me...an amp design that was completely unfamiliar and strange. I think I really probably frustrated my friend with my questions and lack of knowledge...so much to learn, always, but even more so then. My friend, over the course of years now has brought me to the understanding that line amp can be powered by lower voltage than +/-35V, it does not have to be regulated, the higher the voltage the greater the headroom, and regulation would help stabilize the headroom relative to variations in the dynamic output of the amp, but was not a critical factor. He's helped understand how the circuit works, and I was able to draw the circuit out. Last Fall is when this all came to a head and at that point I decided I wanted to find a way to modify the PS-520 supply to provide that unregulated +/-35V power as designed. Concurrent with all the discussion and investigative work in the Fall, I was trolling eBay like I do, and saw a PCB assembly for sale that looked familiar...like exactly like the output line amp boards in the Control Module. It was out of an M-50 console, the much tamed derivative of the M-__ that actually made it to market. These are the boards I talked about in my post a couple posts back that I bought and received yesterday. Anyway, it was a really shocking thing...so weird to see part of the M-__ on eBay. That led me to, with help from folks here, track down the schematics for the M-50 and sure enough the schematic for the M-50 line amp looked suspiciously like the schematic I'd reverse-engineered from the M-__ line amp. So that helped verify a lot. So, yes, the M-50 and only the M-50 has these ultra-high headroom line amps like the M-__, but unlike the M-__ the M-50 only uses them on the main STEREO L-R outputs...the M-__ uses them on the main outs, the 4 AUX master outs, and the 8 PGM BUSS group outs. So with all that new info and understanding I bought a separate step-down power transformer to mount inside the PS-520 chassis so it could produce the +/-35V power. I purchased one that I thought would output the right voltage, but it resulted in too high a voltage...+/-40V. That was because I didn't understand the rated output voltage of the transformer was at transformer's rated max load. I couldn't find a transformer that produced the right voltage for the nominal load. My friend said he would "bread-board" (build the line amp circuit on proto board) the amp circuit and see what happened when it was powered by +/-40V power...his assumption, based on the components used in the circuit, was it would be fine. He's a busy guy too. Month after month passed...I checked in a couple times to see when he thought he might get to it and also gave him an out to just not do it if he didn't have the time, since there was another option: add a zener diode in series after the main filter cap in each side of the bipolar supply to divide the voltage (reduce the voltage to the mixer) and provide some regulation. Cheap, simple, effective. There are even lands on the power supply board to install the zeners where the old 18V regulators used to be. So he called me the other day and we talked through it and we are going with simple and elegant zener solution...costs about $1. I've ordered the parts. I'm also going to replace the fast-blow 2A fuses with slow-blow. The fast-blow fuses are the stock parts in the PS-520 for that +/-18V supply that powers the BALANCE AMP section of the M-520...less current draw there than the total of the 16 high headroom line amps in the M-__, and plus some time back I increased the main filter caps to 10,000uF, so there is just too much in-rush current at power up for those fast-blow fuses...so I have some of those ordered too. My friend also suggested I add a second set of small-value HF bypass caps up in the mixer frame where the +/-35V power traces branch off to each module, to shunt any noise inducted in the power supply umbilical between the supply and the frame. I'm gonna do that too. So, a near decade long mystery is completely resolved, and once the parts get here the M-__ will be able to perform as designed with its relatively unique semi-discrete ultra-high headroom outputs.
 
Holy block of text batman!

Awesome information. I've been wondering about that power rail, and the warble you had been seeing on it. Keep us posted!
 
Yeah, so I'm not really certain what those issues were all about...the differences in the + and - rail waveforms. At this point I continue to be of the mind it is related to the sub-par scope I'm using. I have a Tek 455, but I think it's a bit tits-up and can't be trusted, so I borrowed a 20MHz scope but I'm not sure I can trust it either. The +/-35V supply is very basic, and I've swapped and/or replaced pretty much all the parts and I still get the disparate waveforms between the +/- rails...and the DMM says they are okay, so for better or worse I'm turning the other cheek to what I'm seeing on the scope and taking the assumption the supply is okay.
 
Every time I go into the room where the M-__ sits, which is nearly every day, the torn open power supply silently glares at me; "Why...why do you just leave me ripped open like this?" It doesn't understand being spread too thin or being burned out. I'd like to be interested in putting it back together but that interest, or the energy to have that interest just hasn't been present for months now. Lately I've been thinking I should just go through the motions and force myself to spend even 30 minutes on it.

I did that this evening.

I got the zener diodes installed on the power supply PCB with the intent of dropping the ultra high headroom line amp power supply from the +/-40V power being produced with the add-on power transformer down to something closer to the specified +/-35V power.

Here are the zeners...the little guys sticking up:

IMG_0020.JPG

IMG_0021.JPG

I got them installed along with the new slo-blo fuses and powered it up without a load to make sure nothing blew up or otherwise caught on fire, and then from there to check and make sure the voltage was correct and clean.

No *BANG* or smoke...The voltage was running at around +/-39V. Grr. And my Fluke meter was reading about 20mV AC on each rail. Grr again.

I hooked up the scope to have a look and I got a solid trigger on a nice AC sine wave at about 120mV on each rail. Grr, grr.

Then something flashed in my memory...power supplies have the tendency to run a little wild without a load. So I pulled all the modules from the frame except for one, hooked up the umbilical, crossed my fingers and hit the power switch. No smoke...no drama. And the DC voltage settled nicely at about +/-36V with the load, and AC voltage dropped to something less than 2mV. Much better. And when I looked at the supply outputs with scope again all I could see when AC coupled was environmental RF noise, and very little amplitude even when zoomed all the way in.

IMG_0025.JPG

So it was time to see if it was passing audio. I injected some music to one of the line inputs and hooked a set of headphones straight to the BUSS OUT jack (the high headroom line amp is the output driver for the BUSS OUT jack when switched to +4dBu output level, and, yes, audio passed, and it's clean, and the noise floor is (subjectively) very good. That was with the level switched to -10dBV. I switched in the line amp (-10dBV to +4dBu boost) and still clean and happy...and that already good noise floor got better by nearly 20dB.

:D

So this is a bit of a historic moment here...this is likely the first time this console has been fed the full compliment of power for which it was designed since it was sitting on the bench in the R&D department in Montebello back in the early 1980s.

The sh1t works.

I loaded another I/O Module...again, no drama, works just like it should. And now we're having fun...I just like pushing the buttons...there are so many nuance features on this console and I like the fact that, somehow (since I've actually used it so little), I know my way around the control surface pretty well in spite of all the bells and whistles. It's a fun mixer to use. :)

It came time to load the Control (master) Module...did that...same thing, no drama...Supply output voltage now down to just under +/-35V and that's with 6 of 16 of the line amps powered. So I'll see where it sits with all of them loaded and maybe swap in some lower voltage zeners to maintain something closer to +/-35V with the full load if there's much more voltage drop...maybe not. Anyway, Control Module works like it should as well, except I had a little setback...I'm only getting the left channel in the headphones when monitoring the the Control Room buss...it took me about 15 minutes, but with all the controls and features on this thing I was able to pretty quickly narrow down the issue to something in between the output of the Control Room buss source select switchrack and the input of the headphone amp. The Studio buss works fine in the headphones and there is proper signal at the output jacks of the Control Module. So, I didn't yet fix it, but it should be fairly easy to find the specific problem. I know my way around the Control Module guts pretty well.

Time to put the power supply back together!
 
One small step for man...one giant leap for...well, something like that. :D

Things like that always seem to go in spurts followed by clogs, then another spurt....
 
Yeah...and now I'm chomping at the bit to reassemble the power supply and troubleshoot that Control Room buss output issue. :eek:
 
Back
Top