Urei 535 upgrade

[The Real MC]

Originally Posted by tdot

eventually replacing the slides (which are quite sticky...)

The UREI 5xx EQs are cheap for a reason. There is no known source for those slidepots in low quantities. Besides finding replacements that are an exact footprint (esp if they have to fit a PC board), they are a weird 20K value that isn't even made anymore.

[tdot]

Originally Posted by The Real MC

The UREI 5xx EQs are cheap for a reason. There is no known source for those slidepots in low quantities. Besides finding replacements that are an exact footprint (esp if they have to fit a PC board), they are a weird 20K value that isn't even made anymore.

UREI 535 - YouTube

They're not Alps...but


Do the other modifications I want to perform at least make sense??

[dandeurloo]

They can sound really good. I have one that has been modded to basically have an API output stage added to it. It sounds cool. Don't ask me how it was done. Someone much smarter then I did it. The sliders would be a great mod to do to it. Maybe if I get some time I will look at that.

[brianroth]

I see no reason for the filter caps to be rated at more than 35 VDC since from your description they only have 22V across them. I'm a bit surprised that the regulators are running hot with only 4V dropped across them. Are the heat sink fins hot, or just the regulators themselves?

Best,

Bri

[tdot]

Originally Posted by brianroth

I see no reason for the filter caps to be rated at more than 35 VDC since from your description they only have 22V across them. I'm a bit surprised that the regulators are running hot with only 4V dropped across them. Are the heat sink fins hot, or just the regulators themselves?

Best,

Bri

Somebody with more experience than me told me higher voltage caps have better ripple rejection...so, I don't know.

The heat sink fins are quite hot...after ~30 minutes they're almost painful. I didn't try actually touching the regulator (I'm assuming it would be hotter...)

[tdot]

Originally Posted by dandeurloo

They can sound really good. I have one that has been modded to basically have an API output stage added to it. It sounds cool. Don't ask me how it was done. Someone much smarter then I did it. The sliders would be a great mod to do to it. Maybe if I get some time I will look at that.

I'm planning on ordering a few different types just to see...I found a 45mm 20k with a detent (which I want), but they have LEDs on them (I'd gladly wire the LEDs up if the sliders actually fit)... if I happen to find anything that actually fits well I'll let you know, but as someone said above, its doubtful I'll find something that fits properly.

[brianroth]

The capacitance value impacts ripple, not the voltage.

As for the regulators, perhaps you could try larger heat sinks. I'm still surprised they are running that hot with 22V in/18V out.

Best,

Bri

[effitall]

Like you I picked one up for cheap. Totally works...But the slider caps are cracked, broken, or missing (which I just gave up and removed all of them). In addition the sliders themselves blow chunks.

One of the biggest problems with the sliders is how long the slider arm is. Way too long. If you pull the board out and grab the sliders at their base and move them, they feel pretty good.

I've thought about making a new slider board, but that will also require a little mechanical re-engineering as the main PCB is supported by the mechanical connection to the slider PCB. My 535 is mounted and in use so I'd need to pick up another one to take some measurements and see about the new board.

As for mods...I dunno...They sound pretty good as is. I picked mine up just because it was cheap. Michael Wagener has professed love for it's predecessor the 530 on guitars. The 535 has different EQ points...But is essentially the same.

Overall...It's a well built and designed unit with quality guts. I opted to just replaced the old electrolytics and, for giggles, the 1n4007 rectifiers with the "ultra-fast" replacements (which I don't feel really did anything for it).

I have a mix up right now with the 535 on the "meaty" guitars. None of my other EQ's (hardware or software) quite worked right on them.

[tdot]

Originally Posted by effitall

Like you I picked one up for cheap. Totally works...But the slider caps are cracked, broken, or missing (which I just gave up and removed all of them). In addition the sliders themselves blow chunks.

One of the biggest problems with the sliders is how long the slider arm is. Way too long. If you pull the board out and grab the sliders at their base and move them, they feel pretty good.

I've thought about making a new slider board, but that will also require a little mechanical re-engineering as the main PCB is supported by the mechanical connection to the slider PCB. My 535 is mounted and in use so I'd need to pick up another one to take some measurements and see about the new board.

As for mods...I dunno...They sound pretty good as is. I picked mine up just because it was cheap. Michael Wagener has professed love for it's predecessor the 530 on guitars. The 535 has different EQ points...But is essentially the same.

Overall...It's a well built and designed unit with quality guts. I opted to just replaced the old electrolytics and, for giggles, the 1n4007 rectifiers with the "ultra-fast" replacements (which I don't feel really did anything for it).

I have a mix up right now with the 535 on the "meaty" guitars. None of my other EQ's (hardware or software) quite worked right on them.

I agree! It does sound pretty nice...its a little grainy (which sometimes I think doesn't sound bad) and a little noisy...mine sits at around -80db...while my great river into my burl B2 sits at around <-110db...but its still acceptable. I'm hoping that replacing the op-amps with something more modern would help this a little.

What op-amps do you have in yours? The user manual says that it *should* have Harrison op-amps, but mine seems to have TL074s...

What electrolytic caps did you replace? If I'm not mistaken, there are only 6 in mine.....

[effitall]

What op-amps do you have in yours? The user manual says that it *should* have Harrison op-amps, but mine seems to have TL074s...

What electrolytic caps did you replace? If I'm not mistaken, there are only 6 in mine.....

Mine has the TLO74 opamps. I wouldn't stress the Harrison opamps that much. Pretty much any quad JFET opamp would be a good choice. I would look at the AD8513. I've used the AD8510,12, and 13 opamps quite often and find them to be quite nice sounding. For the caps...There are two main filter caps on the power input...I hit those, and the ones around the voltage regulators. Since the main filter caps are axial, I used a Sprague something or other there. The rest of the power supply electrolytics were replaced with Nichicon HE moderately upsized, and the coupling caps on the EQ's were replaced with Nichicon KT I think.

Honestly...The difference with the caps and such wasn't that great. I didn't go all Jim Williams on it though. More of just service and upkeep. If I got another 535 I'd consider hot-rodding it with some great opamps and perhaps direct coupling it. I think Jim tends to yank the output transformers and put a line driver circuit in its place. I tend to like transformers myself...But they usually aren't the most neutral choice for coupling.

[tdot]

So I did it! Nothing with the faders yet though...
I replaced all the TL084's with sockets, and replaced the filter cap, cut/drilled/sanded out holes for the Neutrik XLR jacks on the back, popped in the LME49740 and it worked! Almost...
I couldn't figure out what was wrong at first, but I finally realized half the circut is being grounded through the 'input ground' lead, and if its not connected to case/output ground, well, things don't work too well... I guess the original circut, the input/output ground shared the same pin.

Though, there is a small issue - whenever I turn the right channel gain knob over ~50% or the left channel gain knob over ~75%, I get what I believe is oscillation? Should I do anything about it? Or just leave the gain knobs at minimum?

[Matt Syson]

Hi
It's most likely the new op amps, not compensated correctly. The new one are so much 'faster' than the originals you need to alter compensation that was probably never needed with the old types.
If you leave it oscillating it can burn out monitor tweeters and generally mess up the sound in different ways on different gear.
BTW increasing the capacitance value of the main reservoir caps will make the regulators HOTTER as there will be less ripple. 35 Volt working is plenty (if the unregulated is around 20 - 25 Volts) so you don't need 63 Volt types.
Matt S

[tdot]

Originally Posted by Matt Syson

Hi
It's most likely the new op amps, not compensated correctly. The new one are so much 'faster' than the originals you need to alter compensation that was probably never needed with the old types.
If you leave it oscillating it can burn out monitor tweeters and generally mess up the sound in different ways on different gear.
BTW increasing the capacitance value of the main reservoir caps will make the regulators HOTTER as there will be less ripple. 35 Volt working is plenty (if the unregulated is around 20 - 25 Volts) so you don't need 63 Volt types.
Matt S

Well it only starts once gain has reached a certaint level - that's why I'm wondering if I should just leave it alone and keep the gain at lowest.
I would have no idea where to start in actually stabilizing the opamps, plus there's technically 40 of them in total.....
I wonder if it has anything to do with C32 in the schematic on the output opamp which says 'may be omitted' which isn't on my pcb...

[Jim Williams]

Originally Posted by Matt Syson

BTW increasing the capacitance value of the main reservoir caps will make the regulators HOTTER as there will be less ripple. Matt S

? Less ripple means the regs are working less hard. Heat is generated by the amount of output current of the regulator, in idle mode they are not hot at all.

[S2udio]

Originally Posted by tdot

I would have no idea where to start in actually stabilizing the opamps, plus there's technically 40 of them in total.....

Then you need some expert help....
Which you don't seem to be listening too. ?

[The Real MC]

Originally Posted by Matt Syson

BTW increasing the capacitance value of the main reservoir caps will make the regulators HOTTER as there will be less ripple.

But with the added benefit of better protection against brownout - the more ripple there is, the less brownout the regulator can tolerate.

Originally Posted by Jim Williams

? Less ripple means the regs are working less hard. Heat is generated by the amount of output current of the regulator, in idle mode they are not hot at all.

Actually heat is generated by dissipated power, not just current. The power that the regulator has to dissipate is current times the voltage difference between input and output. Ripple has to be analyzed in the AC domain - area under the curve. Power/heat is inversely proportional to ripple.

[tdot]

Originally Posted by S2udio

Then you need some expert help....
Which you don't seem to be listening too. ?

?
I did listen - I put 35v caps in

Anyways, the output noise level with no input did drop from around -80 db to around -114 db, which I can't complain about I don't know if that's more because of the XLR connectors vs the screw on connectors, or the opamps, or both.

I don't think I have the skill to fix the oscillation, but it seems to work perfectly fine if I keep the gain level under half - so I'll probably just leave it alone...

[Matt Syson]

Hi
The worst case heat dissipation for a regulator would be if were fed pure DC with no ripple as it is then simply Ohms law being the voltage across it multiplied by the current.
If there is ripple then some of the time there will be reduced dissipation. You would have to analyse the situation fully to find out exact dissipation. It depends what your meter is actually reading when you measure DC + ripple.
Increasing the capacitor before the regulator will reduce ripple and increase the 'average' DC level, therefore making the reg hotter.
Matt S