[makasin]

Can the SOHA be modified to work as a power pre-amp?

 

[dBel84]

I was using mine as a preamp for my 41Hz AMP6 up until recently. I had built it with this dual purpose in mind and thus it had beefier supply caps for the O/P, TO-220 regs (with heatsinks) and a buffered output stage for higher current. It worked pretty well too but I nearly blew my amp once by forgetting to power up/down in the correct sequence. Luckily fuses are easy to replace..dB

 

[makasin]

damn, that sounds like something I would be interested in building. One of my friends coworkers wants me to build him a preamp. How did you get buffered outputs? Is there a site with the design or something?

 

[tomb]

dBel is famous as the first SOHA builder to add BUF634's - two TO-220's in the feedback loop of the opamp. As he states, however, you need to incorporate a delay in the SOHA. There's quite a bit of DC that shoots through at power up - until the opamp catches up and zero's the offset.

 

[cfcubed]

Hey, I'm a noob at this particular stuff & building my 1st SOHA considering this too. To cope w/the power on problems I ordered up parts for this simple power on delay circuit (w/mouser 551-EC2-12NJ relay to make/break output lines). Hoping it might suffice over the more expensive proper way to do it

 

[dBel84]

famous???? where's the paparazzi

. Have a look at Runeight's implementation of the OPA-BUF634, this is what I did on a homemade drop-in adapter. I think Steinchen's JISBOS buffer is the way to go though, he will be finalising his prototypes when he gets the time again and is then going to set up a group buy for pcb's. Following his recommended modifications to the build will be all that is needed.

As for the output dc protection - I would strongly encourage the use of AMB's epsilon as this would protect for both power up and down surge, while the delay circuit would only apply to power up (if enough delay is allowed for >15sec). I can't remember the exact voltages but they were >10V at the output with either powering up or down..dB

 

[tomb]

Quote:

Originally Posted by dBel84 /img/forum/go_quote.gif famous???? where's the paparazzi . Have a look at Runeight's implementation of the OPA-BUF634, this is what I did on a homemade drop-in adapter. I think Steinchen's JISBOS buffer is the way to go though, he will be finalising his prototypes when he gets the time again and is then going to set up a group buy for pcb's. Following his recommended modifications to the build will be all that is needed. As for the output dc protection - I would strongly encourage the use of AMB's epsilon as this would protect for both power up and down surge, while the delay circuit would only apply to power up (if enough delay is allowed for >15sec). I can't remember the exact voltages but they were >10V at the output with either powering up or down..dB

"Famous"? - yes, well - I need to tone down my hyperbolic references lately. Although, I'm wondering if ">10V" is hyperbolic enough for another related discussion.

Nevertheless, don't minimize your contribution as the first to try the BUF634's on the SOHA.

 

[regal]

Doesn't the SOHA have too much gain to be used as a preamp?

 

[cfcubed]

Quote:

Originally Posted by dBel84 /img/forum/go_quote.gif <snip> As for the output dc protection - I would strongly encourage the use of AMB's epsilon as this would protect for both power up and down surge, while the delay circuit would only apply to power up (if enough delay is allowed for >15sec). I can't remember the exact voltages but they were >10V at the output with either powering up or down..dB

Yes, that's why I said the "epsilon 12" was the proper way.

Maybe I'm a little dense, but could someone explain how the relay circuit I referenced, with a non-latching relay, doesn't block power down surges? If its wired to the SOHA's source 12v (maybe w/pot to tweak relay circuit input V to just cause latch) & the relay has a 1ms release time, wouldn't the relay release (breaking output +s) at power-off happen before any surge could get across? The relay delay time should be able to be increased > 7 secs w/diff cap values (but who knows how far)...

If the power off wasn't orderly enough, how bout a DPST power switch w/1 leg for SOHA power & other for relay circuit power?

I wasn't going to commit my SOHA to this simple relay w/o testing it first, but maybe some solid info could save me the time of doing that

> (regal) Doesn't the SOHA have too much gain to be used as a preamp?

Couldn't you play with the output resistor values to compensate for this or would impedance presented to the amp then be a problem? I'm SIP-socketing those resistors so I could play w/their value & use as the relay make/break points.. Personally, I was hoping for enough gain to support a tone stack after the tube if I want it, but that's a stretch in a few ways...

 

[dBel84]

Quote:

Originally Posted by cfcubed /img/forum/go_quote.gif Maybe I'm a little dense, but could someone explain how the relay circuit I referenced, with a non-latching relay, doesn't block power down surges?

no, you're not the one being dense here

, if the coil supply bleeds instantaneously, the relay will disconnect the source. You wisely said you would make sure it does what it is meant to do before use. It won't protect against component failure which the epsilon still will, but the likelihood of that happening is much smaller.

The gain of the SOHA is approximately 12, which is high for a preamp. I guess this comes down to why you are building it. I wanted a headphone amplifier that I could use on my desk and optionally route the signal to an amp in my lab to drive speakers. The pre-amp component was secondary and worked adequately with the amp6. I would not choose the SOHA as a dedicated preamp in my main system as there are designs that suite this need far better.IMHO..dB

 

[cfcubed]

Quote:

Originally Posted by dBel84 /img/forum/go_quote.gif no, you're not the one being dense here<snip>

Well unless someone posts here that the relay thing I described certainly doesn't work, I'll give it a try & post back what I find. I'll try the 12v SOHA supply collapse 1st just to see, then goto the DPST since I'm using one.

Quote:

Originally Posted by dBel84 /img/forum/go_quote.gif The gain of the SOHA is approximately 12, which is high for a preamp. I guess this comes down to why you are building it. I wanted a headphone amplifier that I could use on my desk and optionally route the signal to an amp in my lab to drive speakers. The pre-amp component was secondary and worked adequately with the amp6. I would not choose the SOHA as a dedicated preamp in my main system as there are designs that suite this need far better.IMHO..dB

My purposes too, primarily a headphone amp, maybe a preamp (feeding an amp not too precious to me or powered speakers). The pwr on/off surge protection is to protect my headphones... I *know* I'd be forgetting to unplug them.

 

[cfcubed]

An update: Measurements of my SOHA outputs confirms it takes about 10 secs to settle down (< +/- .5? VDC) and its 12v supply takes about 1 sec to discharge, so will be going w/the DPST method to cut SOHA & relay circuit pwr. When I get to making up the relay delay (& maybe tone stack) I guess I'll post to a new thread like "SOHA non-buffer mods" or something.

 

[cfcubed]

OK, done testing the relay circuit... Using the circuit I ref above w/a 220uf cap (in place of 100uf in schem) yields a 15-17 second delay as long as you don't short cycle the power on the SOHA... Good not to turn things off then on quickly anyway. The DPST route fails because it introduces AC hum by brining the SOAH's 12v in proximity to AC within the switch, but circuit works well for both on delay & off disconnect just directly connected to SOHA's 12v(drawing only 12ma when latched). If you try that circuit external to the SOHA's 12v, be sure to include the 10k (approx) bleed-down resistor.

But, learned yesterday that its easy to build "scaled down" epsilon 12s that includes only the delay, in fact thats what I see in most all the pics of builds, and that would be the way to go to avoid breadboarding. You could also just extract the delay portion of epsilon 12 & use that. For me, not using the a PCB has the advantage that I can build the delay long & thin to tuck between the SOHA PCP & my front panel.

 

[dBel84]

Good to hear the relay works as expected, I never thought to mention that several people have modified / slimmed down the e12 to suite specific space / other needs.



[One thing regarding the hum - it is possible that this is originating from the relay coil ( I tried to find the reference for this but don't seem to have it bookmarked. Just something to bear in mind.) If it is AC, you could break the B+ track on the board (after the CCS) and have the relay switch the B+ thereby eliminating AC interference. ] scrap that thought - I left it in in case you ever use the relay to control B+, but this is a VERY different application , my mind has a habit of wandering, no regard for the sod at the other end

..dB

 

[cfcubed]

Quote:

Originally Posted by dBel84 /img/forum/go_quote.gif Good to hear the relay works as expected, I never thought to mention that several people have modified / slimmed down the e12 to suite specific space / other needs.

Yeah, I'm not new to electronics but new to this area & didn't do enough searches re: e12 to find someone else's "slimmed down" version... But, as I say, this circuit is working very well in the SOHA...

Re: hum - I guess I should have been clearer... After removing the interconnect w/the SOHA's DPST switch (going direct to board's 12v grd/+ around C7) it's back to absolutely no hum. Followed amb's Wiring & ground info, virtual grounding pot, etc. & very happy w/results.

Again, I guess I'll take a pic of this completed delay circuit & info from my tone stack fun & should create a new thread (to solicit comments

.