Ampex ATR102 PSU current

[damianschwartz]

Hello! i'm going to order some xformers from UTM and given the chance, i asked him to wound a custiom power transformer for my Ampex ATR102, the trafo went dead some time ago, it used to vibrate a lot, and lastly it died. I managed to get it out of the housing, which was filled with some kind of Tar. We know it's +-20 +-20 +5 and the total consumption, but i can't find the current for each of the secondary voltages. I can try to measure the cable gauge and the core size, i think i can manage to do that. But, maybe someone has the answer already??

Thanks in advance

 

[radardoug]

I recently had the power transformer die in my Lexicon 480. I figured out the manufacturer, and found them still in business. Emailed them and asked about getting a transformer. They responded and built me a new one, same spec as the old one. Maybe you could do the same with the Ampex transformer.

 

[Bo Deadly]

Use a bench supply or whatever you have available to test. Find a series resistor to measure the voltage drop across and calculate I = V / R.

Then skip the transformer and make a proper SMPS supply.

 

[Brian Roth]

Then skip the transformer and make a proper SMPS supply.

Wow! That opens up a huge can of worms. The ATR machines already have variable amounts of "out of band" (ie, above 20 kHz) from things such as the front panel transport control panel/tape position readout display. There are 9600 HZ square waves (and multiples) doing capstan servo speeds, also. Since these machines are 50-ish years old now, there's plenty of "OEM digital hash" slowly leaking into the audio circuits.

I currently maintain ten ATR-100's in one facility and see all sorts of digital hash oozing where it isn't wanted. Many times, some of that crap can be reduced by simple things like loosening/retightening screws inside the chassis.

More than a few times via the Ampex mailing list, I see folks suggesting that the "module" with the transport controls/tape position readout be physically removed from the front of the transport top panel.

I guess if a SMPS PSU was substituted, it should be physically located several feet away from the deck so as not to introduce additional "radiated" noise into an already noisy environment.

Ampex had to do some "heroic" design choices to minimize digital hash such as using wide strips of flat braided/grounding "cable" to the head block and motors.

But, they are great machines! Just getting a bit long in the tooth.

Bri

 

[Bo Deadly]

I guess if a SMPS PSU was substituted, it should be physically located several feet away from the deck so as not to introduce additional "radiated" noise into an already noisy environment.

I know you're a expert on these machines but I really sincerely doubt that would be necessary. As long as it was properly implemented and not close to the heads or the reproduce amps, it would be fine tucked into an inside corner. I would not be surprised at all if some noise levels when down actually (I would be surprised if it didn't). Especially mains noise.

 

[VertigoRecording]

I know you're a expert on these machines but I really sincerely doubt that would be necessary. As long as it was properly implemented and not close to the heads or the reproduce amps, it would be fine tucked into an inside corner. I would not be surprised at all if some noise levels when down actually (I would be surprised if it didn't). Especially mains noise.

I would have to strongly disagree with Bo Deadly's comment since the machine uses 28kHz to modulate the tension arms. Any beating or demultiplexing of that frequency would cause complete havoc. Same goes for bias and erase frequencies which are in the hundreds of kilohertz. Think about it.

 

[Bo Deadly]

I would have to strongly disagree with Bo Deadly's comment since the machine uses 28kHz to modulate the tension arms. Any beating or demultiplexing of that frequency would cause complete havoc. Same goes for bias and erase frequencies which are in the hundreds of kilohertz. Think about it.

The only scenario where a properly implemented SMPS can bleed into a circuit is if it's sourcing high current and it's in close proximity to either magnetics or high impedance / high gain circuitry. The proximity required is proportional to the current. There is enough space inside that I don't think there would be any negative effects. I've never tried it of course so there's no way to know but there is no evidence to justify immediately dismissing the possibility of using an SMPS inside the ATR 201 chassis. My guess would be that it would work well. But you can't just slap some SMPS inside and wire it up. There needs to be a PCB with the right filtering.

 

[VertigoRecording]

The only scenario where a properly implemented SMPS can bleed into a circuit is if it's sourcing high current and it's in close proximity to either magnetics or high impedance / high gain circuitry. The proximity required is proportional to the current. There is enough space inside that I don't think there would be any negative effects. I've never tried it of course so there's no way to know but there is no evidence to justify immediately dismissing the possibility of using an SMPS inside the ATR 201 chassis. My guess would be that it would work well. But you can't just slap some SMPS inside and wire it up. There needs to be a PCB with the right filtering.

I would have to agree with you. A linear supply would work 100% of the time, the switching supply's effectiveness is anybody's guess.

 

[radardoug]

Plus why redesign the existing supply? There is no market for ATR100 supplies.