roland 808 drum machine u.s power conversion

[cod65]

VA and watt is indeed not entirely the same, but in this case it is going to be negligible.

Practice and theory are 2 different things. They're close but the difference is 'experience' and 'wrong information'. Let me explain with this example: The schematic says that between yellow and yellow there is 10 VDC, well good luck with that, if a transformer has DC on its coils then it is more a magnet than a transformer.

The given is that the 2 existing 2 x 3VA transformers make it work. Chances are this is enough power. If you want to make sure then you have to measure. I think this is a good exercise for you because it will give you certitude and explain the theory in practice and build some experience

What do you measure:
First
A/ AC Voltage of each secondary winding in operation, with the synth powered on and preferably making music.
B/ DC Voltage of each secondary winding. Obviously this should be very near to 0, but just in case there is new physics out there or a very big issue that needs repair
C/ AC Voltage of each secondary winding 'open' (= not connected to the synth, so you have to disconnect the wires)
By comparing this you can see how much of a difference the load the machine gives on the transformer. If the voltages in A are dropping heavily compared to C then the transformer is not big enough. I suspect this will not happen as there are no complaints now.

Second
- AC current on each winding while playing full blown music.
Now you can multiply this current with the above voltage in A and you know how much it really needs in VA. You can size the transformers accordingly. Here you indeed take some margin . In this case 'the biggest you can fit 'comfortably' in the case. Else as this is a small power supply, 150% is just a tad overkill but why not. If it was a very big power supply then maybe 110% is already enough.

brilliant. I cant thank you enough for your time..... means a lot to me!

I hear you on 'experience' tip, but I am way too intimidated by this all-metal class 1 chassis to run it like that with "hood open" -the top lid section, which contains the majority of post power supply circuitry, can only lay upside down when it is open - I would have to work controls from underneath somehow- very awkward  and scary !    I hear your testing plan and I think I understand the theory, but I'm way too intimidated to run this thing open like that - I will be happy if I can disconnect the top circuits isolate the power supply and verify clean 5 volts and 15 volts dc..... if i can do that I'm good!

thank you though.

[Swake]

the dual regulator TA7179P will work between VIN(18-30v), so I assume after rectification my 18vac secondary should be what .... about 24vdc ?

Yes.
18 x SRQ(2) = 25,4 V
Minus the bridge rectifier. So take 2 diode junctions int consideration = 1 x 0.6 V less -> 24.2

That lower voltage will make it, and Q1/Q2, run somewhat cooler too.

[cod65]

yeah, i was weirded out by the secondaries listed in AC also- thought maybe they were talking about after rectification, or peak or something, but i didn't say anything because i didn't want to seem like a dummy 

Maybe its just a translation error from Japanese

[cod65]

of course another presumption here is that the person who did the mod before me, and put these transformers in place of original, chose the right capacities .

So ... come to think on it further- why isn't my mentor correct- that if i have 500ma fuses and only 150ma-300ma max capacities - why would  want to potentially allow my transformers to blow before  my fuses?  or is it just they are less susceptible to blowing under great surges (transformers are more analogous to 'slow blow' by their physical nature ?)

I read up on this , and it seems that transformers are more robust in their repsonse to big in-rushes, can handle brief peaks much better than glass fuse could - that makes sense why the fuse would be above its capacity- its mainly a matter of how long it can survive an overload current, right ?

Then why would my mentor, a knowledgeable guy, insist that i wanted to have a transformer that was rated a good amount ABOVE the fuse ?  I need to understand this.

[Swake]

Choosing transformers and fuses are 2 different things. While there will obviously be a relation as both are connected to the same circuit, you will find very quickly that there is not a one rule fits all. The particularities of the circuit to be protected and powered as well as its operating conditions are important to consider when choosing either.

A short insight in how to choose a transformer: https://tameson.com/pages/transformer-calculator.

Fuses: https://www.digikey.com/en/articles/fuse-tutorial

Note that the schematic shows primary side and secondary side fuses. It's is a good thing, actually a must, to have both as these have different goals. Protecting the transformer vs protecting the circuit and of course both are also helping avoid your house burning down which is the often forgotten main reason why they have to be implemented.

The table specifies 250 mA AT fuses. Note the 'T' that comes from 'träge', the German word for 'slow' -> they want you to use a slow blow fuse.

[cod65]

Choosing transformers and fuses are 2 different things. While there will obviously be a relation as both are connected to the same circuit, you will find very quickly that there is not a one rule fits all. The particularities of the circuit to be protected and powered as well as its operating conditions are important to consider when choosing either.

A short insight in how to choose a transformer: https://tameson.com/pages/transformer-calculator.

Fuses: https://www.digikey.com/en/articles/fuse-tutorial

Note that the schematic shows primary side and secondary side fuses. It's is a good thing, actually a must, to have both as these have different goals. Protecting the transformer vs protecting the circuit and of course both are also helping avoid your house burning down which is the often forgotten main reason why they have to be implemented.

The table specifies 250 mA AT fuses. Note the 'T' that comes from 'träge', the German word for 'slow' -> they want you to use a slow blow fuse.

thanks again, informative articles.

Regarding in-rush current and this design: seems like it's not a concern of the designers with a .250(EU) or .500(US) ma fuse?  I also note that the N.American fuses of .500 in table make mention of a n SGA designation, of which I have not found any info as of yet. All the examples Ive seen so far seem to be common glass fuses or fusible resitors.

I'm  wondering if the later production numbers(of which this appears to be one) switched to fusible resistors in place of the secondary glass fuses because they were having problems with surge/transients? they are effectively 'slo-bow' by design are they not ?


This guy works on a N. America rebuild that contains the  fusible resistors at F2,F3,F4 (although the original schematic I included from the manual at top of thread notes only F2-F3  ... that makes less sense to me and wonder if its a Roland  typo - why would F4 not also be a fusible resistor like this video ? (also I note that his resistors don't look particularly special - I wouldn't know if these were actually 'fusible resistors' by visual, myself)

https://youtu.be/e6lGCrhv1kE?si=KL_HkdkyXIy_o8fE&t=127


Here's another thing I've noted with this particular 808 power supply - safety capacitor C1 is entirely absent. I read that its purpose is primarily to stop this machine from putting noise back out into the local mains and show up on other devices nearby- that it will function fine without it (but it seems desirable for that reason alone) - also I read that it will help protect against transients,lightning ,etc.  But I read that it actually magnifies inrush current ?   

The particular device is listed as ECQ-UC1A safety capacitor (looks like it says 473mfd, but I'm assuming that is .473mfd because nothing like that seems to exist then or now)

is something like this a modern equivalent, say .68mfd? Is this necessary/advantageous or should I just leave circuit as is,  without ? How will this affect the fuses ability to withstand any surges or spikes?

https://www.digikey.com/en/product-highlight/p/panasonic/ecq-ua-series

[Swake]

A fuse before a transformer is nearly always slow blow. 220V x 0.25A is 65W, that is like 5 times more than the summed rating of both transformers. -> They took care of inrush.


The switch to fusible resistors probably was because a pick and place machine could place them automatically, much more difficult with glass fuses and that at the end it was cheaper buying one resistor than a holder + a fuse. If you have normal glass fuses, keep it like that, good enough and easier to replace. + you can very easily measure current with a multimeter in that spot.


While there are some particularly good educational video's out there, always consider that Youtube is a source of entertainment The resistors in your linked video are unlikely to be fusible resistors. That really is not going to help for anything else than heating.


Safety caps. Yeah that's another nerds topic  not a must have to make it work and test the device but a must have to respect the safety standards.
Over simplified: The fuse is about 'current'. The safety cap is about 'voltage' and 'noise on the line'. My personal practice in this specific case would be to ignore it (2 classic transformers are not going to make a lot of noise) or if that instrument is of particular value to add a pre-made inline EMI filter module.

Note that it is written 473MC and 473MF. When there is a number on a cap like this, it is expressed in picofarad. in this case 47000 picofarad or 47 nF. The M probably stands for a certain type (I don't know which one, at that time the currently used classification did not exist yet I believe) and the C and F probably stands for the voltage rating. The modern equivalent of what you need is an X2 (or X1 or Y2 or Y1 as these are all ok in this case) at a voltage rating of lets say 250VAC or higher. in short 47nF/250VAC/X2.

You wrote 473mfd. And that is not the same. mfd markings on older capacitors most of the time mean microfarad, note that it is written in all lower case but is sometimes written in all upper case. The use of the 'm' here is extra confusing of course, therefor I always insist on using only the official SI abbreviations : m = milli ; µ = micro ; n = nano; p = pico.




The other delicate part is that the placement of the dot. for some reason they prefer writing things like .47 over 470n and this is also prone to confusion. Have seen many times the dot was missing.

[cod65]

So I got the transformers in there, put on variac and iso tranformer, measured the voltages and put everything back together and its running well. I didn't use slow blow fuses but none of them have popped (yet). I was pricing slowblo at digikey and they were like 7$ each and up. I'm glad I didn't put them in at first because i had mistakenly plugged the 3rd party midi card ribbon cable in wrong and it caused a short and blew F2 a couple times before I figured out what was going on .  Now all is well.

Hers some pics of my layout for any critique or suggestions that might be gleaned from a visual.

I don't like that these transformers have primaries on bottom instead of top,  so close to chassis, but with shrink tubing maybe that is a good thing.
the transformers  have lock washers but i might put a dab of thread lock on the nuts as well

[Swake]

Good job. Seems like you could reuse the same holes in the chassis, that is nice and clean.

If you want the heat shrink to fit more tightly around the cable on the plug you could add a short piece of smaller gauge heat shrink first just on the cable to make it 'ticker'. But like it is now it is good enough.

Those stickers on the trans are weird. I wouldn't be able to tell you if it is 36V or 2x36V while in fact it is 2x18V.

For 7$ you can but an assortment of fuses.... here an example; https://www.aliexpress.com/item/1005004114014014.html
Don't let you blow away by those telling things like "you have to buy these fuses from a reputable source" before you have read or watched real live test results.

[Haenk]

Nicely done, if it wasn't for the "Made in China" and the date code, the modification would pass as "factory", IMHO.

[pope]

Ironically I have an 808 that runs at 120V and I keep saying that I'll change it to 220V but I can't be arsed. It's been running with a step up transformer all these years.

Perhaps we could swap the transformers 

[cod65]

Sure, but these are not the original that i took out, just 2 generic 230vac. You might as well buy new 230. If you have original 115 probably worth some money