Roland vg-8

[Amraks]

Anyone got any idea what voltages these transformers run this was 117v transformer plugged into Australian mains power.

2 red wires are center tapped.

I'm just trying to get a understanding of the specs I should go before getting a custom r-core built via China.

I don't really trust what is written on the transformer to what I see in the vg-88 schematic which uses the same part number.

The voltages here seem a bit low in comparison to what people have measured else where.

Filter caps are 16v
And 2 other filter caps are 25v

What amperage should each leg have on the transformer.

[stj]

20v-9v-0v-9v-20v

open it - follow the primary wires incase it has a thermal fuse under the tape - it could save a lot of work.
btw, the caps are under-rated when you remember that rectified ac is x1.41
so it's gonna be 28v and 13v more or less

[CapLeaker]

I kinda have a hard time believing that the transformer works but lost a few volt on the secondary. Just make sure your settings on the DMM are correctly, A transformer is never correct, because it is not regulated to start with. Whatever AC input voltage is affects the output voltage. Also the load on the output will effect the output voltage. Transformers are just ball park figures.

Let's say the transformer is putting out 19.2VAC. So 19.2VAC x 1.414 = 27.15VDC. The DC is always higher than the AC voltage due to the rectifier.
I don't think there is something wrong with the primary winding, because the fuse didn't blow.
Just unplug the transformer secondary from the board and measure the AC values.
Since your country is 240VAC, the transformer put out twice as much voltage than it should. Probably all the caps are junk. Maybe the rectifier diodes and other components are bad and lugging the transformer down.
Another thing you could do is measure current, or compare the AC voltage on the secondary plugged in vs not plugged in. Should there be a difference, yes… but noting mayor like 10VAC. Lol

[redwire]

I am really surprised there is no fuse board. Inside the transformer the thermal fuse better be there, although the transformer has no safety cert on it which is scary.
Can you give the part numbers on the three guys on heatsinks and on the DIP-8. It looks quite different than the VG88 board (sch attached).

My math,
19.2VCT gives +/15.1VDC raw at no load, +/12.4VDC raw at full load, which is a bit weird as feed to the linear regulators. It's too low for even +/-12V regulators (rails).
8.9VCT gives +/12.1VDC raw at no load, +/10.2VDC raw at full load.
So I also think the transformer label is suspect.

For +/-15V regulators I've used 25VCT and it would drop out and give hum if the mains voltage sagged at some venues, so I use 28VCT with no problems which gives +/-18.6-21.6V into the voltage regulators and there is headroom for low line.

[Amraks]

I kinda have a hard time believing that the transformer works but lost a few volt on the secondary. Just make sure your settings on the DMM are correctly, A transformer is never correct, because it is not regulated to start with. Whatever AC input voltage is affects the output voltage. Also the load on the output will effect the output voltage. Transformers are just ball park figures.

Let's say the transformer is putting out 19.2VAC. So 19.2VAC x 1.414 = 27.15VDC. The DC is always higher than the AC voltage due to the rectifier.
I don't think there is something wrong with the primary winding, because the fuse didn't blow.
Just unplug the transformer secondary from the board and measure the AC values.
Since your country is 240VAC, the transformer put out twice as much voltage than it should. Probably all the caps are junk. Maybe the rectifier diodes and other components are bad and lugging the transformer down.
Another thing you could do is measure current, or compare the AC voltage on the secondary plugged in vs not plugged in. Should there be a difference, yes… but noting mayor like 10VAC. Lol

Hey I haven't tested the transformer as I don't have a step down converter for us main to aus mains.but its got a nice burnt smell.

[Amraks]

I am really surprised there is no fuse board. Inside the transformer the thermal fuse better be there, although the transformer has no safety cert on it which is scary.
Can you give the part numbers on the three guys on heatsinks and on the DIP-8. It looks quite different than the VG88 board (sch attached).

My math,
19.2VCT gives +/15.1VDC raw at no load, +/12.4VDC raw at full load, which is a bit weird as feed to the linear regulators. It's too low for even +/-12V regulators (rails).
8.9VCT gives +/12.1VDC raw at no load, +/10.2VDC raw at full load.
So I also think the transformer label is suspect.

For +/-15V regulators I've used 25VCT and it would drop out and give hum if the mains voltage sagged at some venues, so I use 28VCT with no problems which gives +/-18.6-21.6V into the voltage regulators and there is headroom for low line.

Would this design be sufficient

• Primary:
  • 230V
• Secondary:
  • 28VCT (14V-0V-14V) for the main rails to feed ±12V or ±15V regulators.
  • 9V-0V
• Power Rating:
  • 40-50VA to ensure enough power for all circuits.

[CapLeaker]

Umm… don't need a step down converter. Under no load, if you put 240V into a 120V transformer, you get double the voltage out. Just don't hook that secondary up to anything and just measure it with your DMM. If you plug in the primary in to power mains and the mains fuse blows, the primary winding is shorted.
Question: why do you want to replace this transformer with a different custom one? Maybe convert it and put a switching PSU in? That's what I did before.

[redwire]

I think the old label is the target DC voltages they wanted, not the secondary AC ratings.
Is mains in Australia 240V+ now or did they mandate 230V or something I thought I heard? That can be a 5% increase.
If the voltages are high, it means the voltage regulator IC's will run hotter. Bigger, oversized transformers with a light load on them put out higher voltage, but smaller transformers have crappy regulation they can go +20% at no load to spec at full load.

For the new transformer spec, the 14V-0-14V I think is good for supplying the +/-15V regs and 8.5-9VAC for the other rail.
The green filter caps are surely damaged and need to be replaced. I hope the IC's voltage regulators survived.

If you used a (medical, no Y-caps) EMI filter then regular transformer(s) not R-core could work fine. Roland's goal is super quiet DC power for the audio sections. You could use two stock E-I transformers.

[stj]

most places never went to 230v,
that would cost money, they just say they did it and they are within the 5% tolerance.
so 230v +-5% is 218.5 - 241.5 - sneaky bastards

[Amraks]

Umm… don't need a step down converter. Under no load, if you put 240V into a 120V transformer, you get double the voltage out. Just don't hook that secondary up to anything and just measure it with your DMM. If you plug in the primary in to power mains and the mains fuse blows, the primary winding is shorted.
Question: why do you want to replace this transformer with a different custom one? Maybe convert it and put a switching PSU in? That's what I did before.

Sadly that secondary is gone.

[CapLeaker]

Dang. That sux balls.

[momaka]

Sadly that secondary is gone.

Interesting.
So the secondary side has gone open-circuit and the primary is still fine?
If that's the case, I suspect there may be a thermal fuse on the secondary side that has also gone open. Normally thermal fuses are on the primary side... but I had a subwoofer TX that had it on the secondary only, and the primary was protected via external fuse.

Get a multimeter and try measuring the resistances of the primary and secondary sides.
It's OK if the secondary is reading low-resistance - that's normal due to the thickness of the wire and the (relatively) low number of windings.
Primary may read anywhere from 10's of Ohms to a KOhm or two, depending on the thickness of the wire used for it (and number of turns.) If it's open, that's a problem.

Umm… don't need a step down converter. Under no load, if you put 240V into a 120V transformer, you get double the voltage out.

Not true in all cases. In fact, this is very rare.

Most line transformers meant for 110/120V AC operation have a specific impedance at that voltage input (and frequency). When you double the input voltage, the core often saturates and then the primary side impedance typically drops as low as just the resistance of the wire in the winding - i.e. anywhere from a kOhm or two to mere Ohms. But even at 1 KOhm impedance/resistance, that's close to 15 Watts of dissipation for the primary side winding... which if left plugged in at that power level will eventually overheat the winding and melt the insulating coating on it. My experience with this so far has been that most transformers meant for 110/120V AC operation will open on the primary almost immediately (either via internal fuse or the wire typically burns off close to the transformer leg.) In fact, someone recently plugged in my US stereo amp into 230V AC outlet, and the same happened - small standby transformer blew its primary side completely open (no fuse onboard or inside the transformer) and the large transformer blew a 5 Amp fuse. Apart from that, the rest of the amp is working fine when I supplied the standby voltage with a power adapter and replaced the fuse for the big traffo.

Now, you can always go the other way around, though - put 110/120V AC into a 220/230/240V AC -rated transformer, and you'll get half the output voltages.

most places never went to 230v,

??
We get 230V here in S/E Europe (yes, I measured it)... though depending on which part of the country you live in (or rather, how the power distribution is done in the city/town), I've seen it drop as low as 200V and as high as 238V. In particular, it can be pretty bad in the countryside as there are only a few large transformer buildings for the whole town, and then 3p 230V AC is ran through overhead lines to all of the houses. The ones that are far away from the traffo can see a lot of variation. Plug in a stick welder or several kW heaters on one phase, and you can watch the lights dim quite a bit on that phase. Thought out microwave was broke the other day when my uncle was doing some welding outside.

My math,
19.2VCT gives +/15.1VDC raw at no load, +/12.4VDC raw at full load, which is a bit weird as feed to the linear regulators. It's too low for even +/-12V regulators (rails).
8.9VCT gives +/12.1VDC raw at no load, +/10.2VDC raw at full load.
So I also think the transformer label is suspect.

Hmmm... how are you getting these values?
19.2V AC is roughly about 27V DC once rectified and smoothed (at no load.) May also be a bit higher at no load, depending on how good/bad the regulation of the transformer is. Small TXs tend to have rather loose reg, though.
Also, the markings on the transformer mean that each red wire to CT is 19.2V and each orange to CT is 8.9V, not 19.2V from red to red. Not sure if that's perhaps non-standard, but a lot of new traffos are labeled this way.

Would this design be sufficient

• Primary:
  • 230V
• Secondary:
  • 28VCT (14V-0V-14V) for the main rails to feed ±12V or ±15V regulators.
  • 9V-0V
• Power Rating:
  • 40-50VA to ensure enough power for all circuits.

Voltage-wise, I think so.
Only not sure on the VA rating, but you can judge this by the size. If the new traffo is about as big as the old one (or larger), then it should be OK to use. What does the back of the Roland vg-8 say in terms of power consumption?

[redwire]

[...]
Hmmm... how are you getting these values?
19.2V AC is roughly about 27V DC once rectified and smoothed (at no load.) May also be a bit higher at no load, depending on how good/bad the regulation of the transformer is. Small TXs tend to have rather loose reg, though.
Also, the markings on the transformer mean that each red wire to CT is 19.2V and each orange to CT is 8.9V, not 19.2V from red to red. Not sure if that's perhaps non-standard, but a lot of new traffos are labeled this way. [...]

The math I do is transformer peak voltage ACV*sqrt(2) minus (bridge rectifier two diode drop) and then +/-10 to 20% for high or low mains and the transformer's regulation.
Once I did a bunch of measurements on 10 transformers, load/no-load, rectifier etc and found it's actually fairly messy. Nothing is really a nice pretty sine-wave. Incoming mains has flat tops, the secondary is a mangled distorted sine-wave. Cheap chinese transformers are minimum copper and steel so the flux density is really high and they arfe just terrible parts. There is a one shop making custonm R-core parts that might be good.
Each diode drop is 1.2V because of the high peak currents. Transformer regulation for 50VA parts is typically 15%, better quality maybe 11-12%.
Another way to look at it is 15V linear regulators need at least 2V extra in, that's 17VDC so 2x is 34VDC minimum needed here and add 10-15% for low line. The VG-88 uses those (15V) regs and even if the VG-8 is +/-12V regs you need minimum 28VDC so I say the transformer label is bogus, doesn't make sense - unless it's the design target DCV. We don't know the IC numbers or schematic.

The secondary going open-circuit is weird, usually the primary thin wire melts. I'm worried what shorted and killed the secondary. I'm surprised the caps did not pop!

[redwire]

I checked and the other model VG-88 uses some LDO's.
5V 2A digital reg PQ05RF21 (4-pin) LDO and 5.5V in minimum.
Second 5V 1.2A reg is an LDO AN7705F and 6V in minimum.
The +/-15V rail is a capacitance multiplier and transistor/zener reg.

Custom R-core manufacturer
https://www.aliexpress.com/item/1005004566890764.html ADD A PRIMARY FUSE LOL. It's that paper they use. One guy was happy with what they make.
https://www.custommag.com/r-core-transformers surely big bucks. Australia has many custom tranny makers.

[CapLeaker]

Not true in all cases. In fact, this is very rare.

Most line transformers meant for 110/120V AC operation have a specific impedance at that voltage input (and frequency). When you double the input voltage, the core often saturates and then the primary side impedance typically drops as low as just the resistance of the wire in the winding - i.e. anywhere from a kOhm or two to mere Ohms. But even at 1 KOhm impedance/resistance, that's close to 15 Watts of dissipation for the primary side winding... which if left plugged in at that power level will eventually overheat the winding and melt the insulating coating on it. My experience with this so far has been that most transformers meant for 110/120V AC operation will open on the primary almost immediately (either via internal fuse or the wire typically burns off close to the transformer leg.) In fact, someone recently plugged in my US stereo amp into 230V AC outlet, and the same happened - small standby transformer blew its primary side completely open (no fuse onboard or inside the transformer) and the large transformer blew a 5 Amp fuse. Apart from that, the rest of the amp is working fine when I supplied the standby voltage with a power adapter and replaced the fuse for the big traffo.

Now, you can always go the other way around, though - put 110/120V AC into a 220/230/240V AC -rated transformer, and you'll get half the output voltages.

I know that too. But for a short dirty test, it works for me perfectly fine. This was meant for a short test no load. Sure, voltage may be a little off from a 120V transformer on 240v, but that's o.k. Gets you in the ballpark anyway,

redwire to me 19.2V rectified is roughly 27V. The center tap is 0 volts and the two outside taps have 19.2VAC. Basically 19.2vac - 0VAC - 19.2VAC. One leg is 180 degrees off the sine wave from the other. Each transistor pair are divided to work on each tap as shown in the picture. I did see this formula you are talking about, but I can't see it working like that here. Unless I am getting old, dumber and grouchy.

It does NOT make any sense having 12.4VDC rectified coming from a fullly loaded tranformer on a 15VDC rail. Might as well put a 9V wall wart there.

[redwire]

What are you going to do with 27VDC, it's not good for anything in the guitar synth. The filter capacitors are only rated 25V and 16V, and 27V would make so much heat for 15V regulators.
So am saying it can't be 19.2-0-19.2VAC it would be too much for anything, which makes me think the label is bogus. If it was 19.2VDC - then that is a good value for feeding the 15V regulators, and their filter caps. Only way I can make sense of it.

Looking more at the PC board, I see the other low voltage winding has no center tap and is isolated from the other center tapped winding. I think it makes 8.9VDC to feed a 5V and maybe 3.3V regulator.

[CapLeaker]

What are you going to do with 27VDC, it's not good for anything in the guitar synth. The filter capacitors are only rated 25V and 16V, and 27V would make so much heat for 15V regulators.
So am saying it can't be 19.2-0-19.2VAC it would be too much for anything, which makes me think the label is bogus. If it was 19.2VDC - then that is a good value for feeding the 15V regulators, and their filter caps. Only way I can make sense of it.

Looking more at the PC board, I see the other low voltage winding has no center tap and is isolated from the other center tapped winding. I think it makes 8.9VDC to feed a 5V and maybe 3.3V regulator.

You think this transformer isn't original and someone fuxed around with it or that labeling is far off? Kinda weird. But I agree what you say,

[stj]

how about using a transformer for the +-15v and then using switching regulator modules to get the lower voltages?
18-0-18 transformers are very common.

[redwire]

OP can measure the old R-core size and use this table for a ballpark VA rating. From Kitamura Kiden Co. catalog. Maybe even hit up the company for more information on the original transformer?
Kitamura KRT-18D-A-002S2, K5F1KO, 22453552C0, Label Pri: 120V, Sec: 19.2V and 8.9V for the search engines.
One trap is using an oversized transformer VA. If you run them under their rated load, surprise! they have higher output voltage so vregs end up running too hot.

I think SMPS modules are too noisy, Roland stayed using linear regs to keep the synth quiet, good noise floor, lots of ferrite beads as well. The toroid inductor and big diode on the power board sorta look like a buck-converter? but I can't see it being used due to the EMI. It might already have a TO-220 converter for the 5V rail there?
We don't know the IC numbers but I would be fine with using a 14 or 15VACx2 1A transformer (for around 600mA DC) plus another transformer 8-8.5VAC 2A.

I noticed the other model synth VG-88 has 35V filter caps (not 25V) and a capacitance multiplier/ripple filter for the 15V rails - so it needs a bit more extra headroom for the voltage losses there- but OP said it's the same transformer used in both models?

[stj]

you could use buck convertors to drive the linear regulator circuit with a 2v overhead.