2" Tape 16 track help?

[iomegaman]

So on further inspection it seems to me I was a bit premature in assuming the labels on the transformer actually applied to IT, the box it is in seems to be made from leftover aluminum parts it is not some OEM enclosure...so the labels on it might not have anything to do with it since the voltage do not match t all...
This thing puts out 3 rails of ~ 18V-ish AC and 1 rail of ~12V AC...these rails all get dumped into some pretty hefty S3720 silicone diodes...which are rated at 85 amps so the idea that the tranx is 1.5 RMS seems a bit off to me.

I'm going to current limit the capacitors on a DC supply and slowly see if I can reform them before I throw caution to the wind and ramp the entire thing up...
The main ~17volt rails goes through 4 diodes and gets dumped into the two 75,000 uF/25 volt caps which in turn go to the bus tree on the edge of the PSU, the leg then going back to the main bus strip on the deck (labels 16-20)...so two of the rails get 1 each huge 75k uF/25v Capacitor...(which are no longer made although there are some still available? the 36DX 75kuF do not show up on Visheys sheet anymore)

I am assuming the surge current on these is 30 volts because thats the way Vishey lists all the 25 volts even though these are not on the list at this denomination.

The 75000uf's Caps each have a 680 Ohm discharge resistor (unknown wattage but the size of a pencil so I'm assuming) with a .1uF 100V Sprague cap for HF across the terminals...


SO here's my question: What would be a fairly safe way to slowly ramp up these large value caps and then check them? I am guessing remove the discharge resistor/HF cap and maybe use my DC power supply in time intervals to charge them up slowly with current limited?

I do not have any kind of meter that can do LCR at 75k...I suppose I cn O-scope them and do the math for leakage...but I am open to suggestions.

 

[EmilFrid]

To test leakage, I've used this method, although I'm not sure as to wether there's an upper limit to the values you can test:
https://product.tdk.com/en/contact/faq/capacitors-0015.html
I would dial up the voltage slowly while measuring.

I've seen 75.000uf 40V caps for around 20$ on Elfa here in Sweden. With screw terminals. If it's impossible or a hassle for you to order from them, I'd be happy to help you out.

Edit: i checked and Elfa doesn't have them in stock again until around 100 days. The brand is Kemet. Farnell has them readily available now (at least the Swedish Farnell site)

 

[iomegaman]

To test leakage, I've used this method, although I'm not sure as to wether there's an upper limit to the values you can test:
https://product.tdk.com/en/contact/faq/capacitors-0015.html
I would dial up the voltage slowly while measuring.

I've seen 75.000uf 40V caps for around 20$ on Elfa here in Sweden. With screw terminals. If it's impossible or a hassle for you to order from them, I'd be happy to help you out.

Edit: i checked and Elfa doesn't have them in stock again until around 100 days. The brand is Kemet. Farnell has them readily available now (at least the Swedish Farnell site)

I have found some of these larger value caps stateside and honestly I think I'll order a couple just to get the donkey up the hill...been dragging my feet a bit of late...the caps are just not the "mil-spec" of the ones in the machine right now...thanks for the link, very helpful!

 

[EmilFrid]

@iomegaman if you've found them stateside, that's great. It is probably a good idea to replace those caps anyway.

I'm very excited about this project of yours.

Great, hope you'll be able to make some measurements!

 

[Brian Roth]

75000 uF seems insane to me. I just checked, and the Crown DC-300A power amp used a pair of 13,500 uF filter caps to supply over 150 Watts/channel into 8 Ohms. One can complain about the "sonics" of those antique amps, but they were a PA system workhorse in the 1970's.

Bri

 

[iomegaman]

75000 uF seems insane to me. I just checked, and the Crown DC-300A power amp used a pair of 13,500 uF filter caps to supply over 150 Watts/channel into 8 Ohms. One can complain about the "sonics" of those antique amps, but they were a PA system workhorse in the 1970's.

Bri

You and me both Brian, these things are $107.89 on Mouser right now...HOWEVER...at this point I'm going to try and and reform them...the transformer in this thing is about 6" X 4" deep torridal with what looks to be AT LEAST 8-10 AWG wiring...the power supply in my Soundcraft is smaller and thats rated at like 300 watts if I'm not mistaken...
I haven't reversed eng'd the audio cards/playback/rec yet but they look to be all discrete with some coils and switches, things like LM301AH's which run on 18v but only suck 1.8mA (x24).

I tried running one of the 17v rails into a dummy load resistor (100w-1-Ohm) via my "floodlamp current limiting" gizmo (Hat tip to Uncle Doug on youtube)and I hit 10 amps in no time and the light went bright...and that just one of 4 rails...

When I do the capacitor math for ripple smoothing at 18v I get no where near 75k uF...BUT there's only one cap per rail, and I'm guessing on current requirements...shooting in the dark with sunglasses on.

 

[Brian Roth]

I haven't been following this thread too closely (apologies) but I somehow think that PSU design is WAAAY over the top in terms of design. Try to set up a lab PSU and make actual measurements for one/three/whatever channels and determine actual current draw.

I don't have the manual here, but a MCI jh-24 was full of 5532 opamps and the PSU was much more reasonable...IMHO.

Bri

 

[uplinkal]

This is obviously going to be a summer project so don't @me...but I am picking up a 2" 24/16 track tape machine that has been sitting for a while...its been in a smoke free studio and there's no telling when the last time it was run...yes I know its a "can-o-worms" but the price is in my "build budget" that I have not used at all this year or last and I want to take it on.

Here are the "knowns":

The machine is a 2" 24 machine with Ampex cards and a 3M transport , it currently has a 3M 16" butterfly head track on it which I have inspected and is in really good shape... has all 24 channels with Ampex cards that each have an input/output transformer on them (there's also inductors on some of the cards)...all cards are there...machine is 99% there with cables and everything the only thing missing is the drive belt...which of course might be a MAJOR problem...it has the 16 VU meter stack as well...

The company is "Bouse Manufacturing distributed by "Audio Concepts Dave Kelsey"...model 1600 apd

Obviously it will need a thorough going through and the biggest issue I am facing is a drive belt...

The worst case scenario is I end up with a decent head and a bunch of Ampex cards and a pile of metal scrap...the optimal scene is I find a way to resurrect it and push back into a more "analog" market just for kicks and giggles...those old capacitors are going to be huge question marks but from my limited research it was built in the 70's there is almost no info on this company (man I seem to be a magnet for old abandoned business goes bust gear from that era, looking at you Amerimex Console)...

Anyway it certainly looks 90% like the 3M M79 machines...the difference is in the control panel and switches...this is obviously a customized version...

So what are the chances there might be someone here who would know or know of a good resource for 3m belts/parts?

I'm headed to San Francisco next week and will try to stop by B.A.S.E. (Bay Area Studio Engineering) as they seem to be the west coast emporium of information for this stuff...

Anyone have any suggestions other than "Don't Do it Iomegaman!!!

Just for kicks and giggles, try Pro Audio LA at Pro Audio Studio Equipment | Pro Audio LA.
They mostly deal with newer high end gear but maybe one of their staff will know how to find what you need. There is also Jay Kingery at RCI Systems in Beltsville Maryland and Greg Lukens (Greg Lukens Audio on Facebook. that's the sum of my contacts for what you need.

 

[iomegaman]

Thanks uplinkal...will check them out...

Just for update the power supply is running fine, only needed to replace 6 caps all smaller "non-mil-spec" and it was just to err on the side of caution...currently making board extenders/audio card jigs so I can test out all the busses and cards...

All of those big caps reformed and held charge within spec.

The belt I think I can measure now...

 

[aaquilato]

Mara Machines…
https://maramachines.com/

 

[EmilFrid]

Thanks uplinkal...will check them out...

Just for update the power supply is running fine, only needed to replace 6 caps all smaller "non-mil-spec" and it was just to err on the side of caution...currently making board extenders/audio card jigs so I can test out all the busses and cards...

All of those big caps reformed and held charge within spec.

The belt I think I can measure now...

Great news about reforming those big caps. One headache (and expense) less. And great that the psu is working. Well done. Looking forward to hearing about the audio cards.
Have you seen anything suspicious or do they all look fine at least cosmetically?

 

[iomegaman]

All of the cards look good cosmetically...I think this machine was plenty-too-much for whoever owned it last...I know the studio in Phx it came from claimed to have: Studer A827 24 track (2) with remote, MCI JH24 24 track (1) with remote, MCI 16 track and at one point a "Stephens 16 track" which I believe is actually this machine...mainly because the meter bridge looks exactly like a Stephens 16 track but those units are pretty rare and there was no mention of a Bouse...which is what this IS...basically I think it sat in the corner of an AC climate controlled studio for 20 years because someone put the wrong fuses in it and there was a broken belt and NO DOCUMENTATION>>>
I think if you add that stuff up it was probably ignored...but as I dig into it it looks to be a very solid piece of kit, all the audio/playback cards have Jensen Input/Output Transformers...everything about it is over-kill...which is a nice problem to have years later trying to bring Lazarus out his tomb...

 

[waltzingbear]

Good to see you moving forward
Yes, reforming was the best path. The $14 cap you referred to was made in 1980, so it would have needed reforming as well. The 36D’s were great caps and generally last a loooooog time.
This thing was obviously over built, but in this case it’s probably a good thing.
But you better get onto finding a belt or it’s all for naught.

 

[EmilFrid]

Isn't there anyone in the world making custom belts? There are standard belts still being made, so the manufacturers are there...

@iomegaman sounds exciting. Could imagine this thing sounding incredible

 

[iomegaman]

Just a quick note on reforming old large capacitors...I spent a lot of time online investigating various methods/inventions/contraptions/etc for reforming old caps...not one of the methods or links had any serious advice on really large value (but lower voltage) capacitors...the ones in this power supply range from 63,000uf /25 volts to 900uf/150 volts...almost everything you are going to find online relates to more "standard" values...smaller caps used in old tube radios and etc, NOT Mil-spec 50 year old caps...but in theory the principle should remain the same...

What I found worked best was this link that CJ referenced in a thread of the same name...
Basically you use a large wattage resistor to limit current and watch the voltage drop across is, and also keep a MM on your cap to see how it is responding to a slow ramp up of voltage at LOW current...I discovered that large capacitor values need a lot more time probably due to extended layers of foil/electrolytic but I was being overtly careful an unhurried...no point in rushing current into a 50 year old device...

I used a 10k/ 1/2 Watt resistor on the negative lead of my bench power supply...turned everything down to zero and powered it on...turned up current until the indicator showed it was getting past the resistor (less than 50uA) and then gradually turned up the voltage until it was 10% of the rated voltage of cap in test...on large value farad caps the voltage drop on the resistor at the beginning of the process is going to be very close to your power supply voltage...ie: in the 23,000uf/30v cap 10 volts in will show almost 8-9volts drop on R but it will slowly start to drop and the voltage on the cap will go up, left alone it will reach 99% of psu voltage...I did everything in small increments usually 10% of rated value and when I was done would check to see how long a cap would hold its rated voltage with no load...all of them held up to 90% overnight...which is a win as far as I'm concerned.

 

[iomegaman]

So I've encountered a strange conundrum here...(understatement of the year)...on the main power supply rack there is a smaller transformer that appears to NOT be hooked up to anything...it has two sets of secondary windings and the primary located on a separate bus strip under the main tranx bus strip...there's a chance it was meant to take power from the 18volt secondary of the main transformer since the bus strips line up accordingly...so on a whim I loaded up my variac dropped it to 18volts and connected to the primary side to see what I got on the secondary...two sets of 2nds both giving me about 60v. AC...I'm going to fish around in my photos and the wiring guide I am creating but for the life of me I cannot figure out what this is for...

Why would I need 60v AC...probably close to 5 amps...?

 

[ruffrecords]

Why would I need 60v AC...probably close to 5 amps...?

Deck motors, solenoids???

Cheers

Ian

 

[iomegaman]

Deck motors, solenoids???

Cheers

Ian

Yeah maybe...but also looking over things I think I might have it backwards the black/white set I considered "secondary" might actually connect to the 110 leg on a bus strip by jumpers...on a whim I ran the variac that way and measured the other side and I'm at a more reasonable 30 volts...I just assumed that 2 sets on one side would both be secondary and the single side would be primary...

There is a brown set of wires that attached to the black/whites that traces over to the main power bus strip...those wires also have a fan connected that runs off of 110 that cools the housing where all the silicon diodes are connected.

Swimming in peanut butter in the dark without a schematic...go slow...ramp stuff up carefully looking for smoke...document until it makes sense...

 

[iomegaman]

Now I enter the dark and mysterious world of "WTF?"...
As I continue to map out each and every wire/connection/capacitor/fuse/diode etc of this power supply I have run into a serious question about what was going on...

The power supply has those huge 36DX caps...as I trace down power and rectification and check my voltage I'm down to the last one which is a 2100uf 40 volt with a recitifier connected to it...the rectifier is supplied with a 15volt ac from main transformer...the rectifier puts out 12 volts DC.
Here's the strange part...I have checked my documentation, and all my "before test/teardown" pictures...the capacitor + side was connected to the rectifier - negative output...whoever had this thing before me, or perhaps someone before that definitely had this polarized capacitor taking 12 volts of negative rail into the positive side of the capacitor...

Here's the pics to prove I'm not insane (mostly)
As you can see I disconnected the rectifier to test...to be fair I did pull all these large caps out when I reformed them, but I also took pictures from top and side prior to that and can see without a doubt that this cap was oriented with the positive as it sits on the photo...(as seen in 2nd photo which was prior to reforming)>>>

(If you zoom into the 2nd photo you can see the black "added" wire with electrical tape on it is going to the negative post of the capacitor...zoom into the rectifier and you will see that THAT is the positive side of the rectifier...I'm not insane.)


I'm wondering if it is just some tech error or what the heck was going on in studio land?


No wonder this thing sat unused for years...not sure if this was a "Bouse" error or some tech after the fact...who knows?

The unfortunate thing here is now I realize that I need to check/double check all power supply rails/rectifiers/diodes and capacitors to make sure things are at minimum properly polarized...there's a quite a few polarized caps in this thing...I've already replaced everything under "normal size" and am quite impressed with how robust this mil-spec 36DX caps are...when I hooked it back up proper the caps there gave me solid 15 volts no load...

 

[EmilFrid]

No good. Sucks that you now might have more problems like this lurking around... maybe a "tech" messing up? Looks like the wires going to the rectifier is a bit melted, which would indicate that someone might've worked around/on the rectifier.