pm-1000 line driving capability

This has been discussed briefly in one of the Edcor transformer threads, but I figured that it is of interest to enough people that it deserves it's own topic. If this has been discussed in depth already, please hunt me down and torture me; I can't think of any topic besides the pm1k that spawns so many unnecessary posts here. So here goes...

Is there enough juice at C37 on a pm1k to drive a 600:600 output happily?
I know almost everyone who racks these things is doing it, but a few respected people here have said it's a bad idea.

We know that hi-z out is fine, but what about those of us who require 600ohm lines (my console has no usable hi-z ins to speak of)? I assume 10k:600 would be ok if you don't mind the dB loss.

Whadayall think?

-Ben

ok thread...you can have one bump, but that's it.

It's hard to get feedback on these things for some reason. Good luck.

Ok, I'll up the ante. I'm about to rack a 2nd pair, this time with the output transformers, but I'd really like to have a direct out (unbalanced) for monitoring also. Can I drive both simultaneously (from~C37) or is that going to cause trouble?

thanks,

Mike

IF C37 can effectively drive a 600:600 trafo, then I don't see why it would be a problem. I always split the output of my pres at the patchbay, so I can monitor while tracking.

However, if it can't drive the 600:600 in the first place then there is a problem. I'm concerned about the the fact that people have complained of the sound of their pm1ks with the output trafo versus those with a hi-z out. Rafafred said in that same thread that the input modules can't do it.

http://www.groupdiy.com/index.php?topic=14173&highlight=edcor

Hopefully we can get some definitive answer on this one. I've got a pile of money with Edcor's name on it.

-ben

OK...

Without analyzing the circuit in depth I would say it would it should be able to take 600 ohm loading without any problems.

BUT - seeing as the C37 is a measly 33uF in value, you'd be well advised to increase that to at least 470uF. preferably 1000uF.

In general, PM1000 used far too many low value coupling capacitors, not good. The accumulated phase shift of all those capacitors will certainly "muddy up" the low end.

[quote author="Brown Note"]IF C37 can effectively drive a 600:600 trafo, then I don't see why it would be a problem. I always split the output of my pres at the patchbay, so I can monitor while tracking.

However, if it can't drive the 600:600 in the first place then there is a problem. I'm concerned about the the fact that people have complained of the sound of their pm1ks with the output trafo versus those with a hi-z out. Rafafred said in that same thread that the input modules can't do it.

http://www.groupdiy.com/index.php?topic=14173&highlight=edcor

Hopefully we can get some definitive answer on this one. I've got a pile of money with Edcor's name on it.

-ben[/quote]

Are you saying the ORIGINAL output xformers won't perform well?

I didn't see Rafafred metioning those, just the Edcors... what is the original output tranny's imp ratio anyway? I've read a lot and do recall anyone saying they didn't like them.

The original are tamuras, if I recall right, and have probably much higher inductance tha EDCORS, so the amplifier should work better with the originals. But still, I would really do a small two transistors buffers, with TO220 packsges for driving 600 ohms transformers with the PM1000. Have alook around. There are PLENTY of buffers like this in the forum. It´s really simple.

I like to drive low impedance transformers with lots of juice, that is.

I did a more controlled test with the 600/600 Edcors, and would say that the difference is less pronounced than I originally noted. My initial tests weren't very controlled, basically me singing/talking into the different configurations and recording the results. Of course, this meant different performances. I repeated the tests with a CD run through a passive direct box and then into the PM1000s and recorded the outputs. I wound up using about 44dB of gain. Again, I heard the same results, but not quite as noticible. I also played the files for my buddy and he picked out the transformerless as being the better sounding file, but we both agreed that it was pretty close, almost negligible.

Here's my 2c on the topic.

I have 8 modules in my console right now, four have direct outs with output trannies stolen from the master modules. I use the master modules whenever possible, because it's a hotter output and sounds better to my ear. How much better? Gosh, I have no idea. When you do the direct out straight of C37 it's not a very hot output. I usually have to turn the gain knob one more notch to get a level near the output modules level. Which, is another reason I use the master modules whenever possible.

That said, I've done some recording, not much, with the output transformers and wouldn't say it sounds bad or worse. Just quieter. I haven't done any critical listening tests though in any depth to see whether it's just quieter or sounds worse. I should probably do that. I'm still kind of finishing up work on the console, which is why I haven't done any in depth tests. All of my modules have C37 at 470uf now I believe. I'll have to try switching it up to 1000uf and see if that makes any difference. I also doubled the value of most of the caps in the modules, aside from those in the eq.

Hmm...I should do some a/b recording out of the xformer output and master output and post them, shouldn't I? Hehe Maybe this weekend I'll fire up a listening test for you to decide for yourself. :guinness: :sam:

I'll need beer....lots of beer...

[quote author="rafafredd"]The original are tamuras, if I recall right, and have probably much higher inductance tha EDCORS, so the amplifier should work better with the originals. But still, I would really do a small two transistors buffers, with TO220 packsges for driving 600 ohms transformers with the PM1000. Have alook around. There are PLENTY of buffers like this in the forum. It´s really simple.

I like to drive low impedance transformers with lots of juice, that is.[/quote]

I'd be interested in trying this...but I'm a complete retard. I need pictures and graphs and flow charts or it takes me two months to do something like this. I'm still an electronics NOOB, for the record.

Are you saying the ORIGINAL output xformers won't perform well?

I didn't see Rafafred metioning those, just the Edcors... what is the original output tranny's imp ratio anyway? I've read a lot and do recall anyone saying they didn't like them.

Click to expand...

According to the schematic the original Tamura output transformers are 600:600 as well. As far as original vs Edcor I thought we were, to a point, comparing apples and apples. C37 wasn't meant to drive an output trafo in the first place. After C37 the signal goes through another amplifier stage in the master section before it hits that transformer.

So that's basically what I'm concerned about. We're probably splitting hairs here, but my drums always go through these things not to mention a good percentage of my overdubs. I don't want to degrade my sound but, I really need a 600ohm output. Any opinions on a 10k:600?

khstudio and digitaldrummer: I'm gonna pick up some edcors soon anyways. I'll a/b trafo out versus hi-z on the scope, and I'll do some listening tests. Someone should do the same with the Tamuras.

-ben

[quote author="Brown Note"]I don't want to degrade my sound but, I really need a 600ohm output. [/quote]

Why do you believe you need 600 ohm output?? :roll:

Why do you believe you need 600 ohm output??

Click to expand...

Correct me if I'm wrong here. I'll give you the lowdown of my setup...

Because of latency issues I don't like to monitor tracking through my audio interface (2X Gadget Labs Wav8/24). Therefore I split the outputs of my mic pres at my patchbay, so I can monitor through my console. The console is a Yamaha m1516 (more or less a 16x4 version of the PM-2000), which only has 600ohm transformer-balanced inputs. I am assuming that, because of the load the pre's see from the console the signal going to "tape" is suffering. Am I way off here?

-ben

Any opinions on a 10k:600?

Click to expand...

Really. That´s not what you need. If you use a transformer on the output, you loose maximum output and gain. That´s really not the way to go. The way to go is building a simple buffer and hearing for yourself if that makes a diference. Just breadboard a simple buffer.

You could include the output buffer in the feedback loop for better results also.

I dont very much about these cosoles, so preface my comments with that-

Seems like these consoles were originally built to a 600 ohm standard. Were there originally direct outputs on the console to begin with, or are you just using an aux or something? If you are just using the diret out as a monitor path, do you *need* to go through the trouble of balancing it with a transformer?

They probably figured out a somewhat decent gain structure for the desk based upon the positive comments these things seem generate. Instead of re-inventing the wheel, why not preserve the original gain structure of the console and use that for your direct outs, its probably not too much work. I would assume that each channel feeds an unbalanced signal to the buss where it meets the output summing/line driving stage and the output transformer. Why not take the unbalanced direct out you want to use and then feed that to a copy of the line driving stage that you can build on a little piece of perfboard inside the box? This way, you know you'll have the might to drive the transformer "properly" to whatever spec they had designed and you'll get the "sound" of the console in your direct outs. I would guess their line driver stage isnt very complex and would proably be easy enough to wire up point to point on some perf.

Anyhow, if it were me and the direct outs were critical enough to me that I would want to use a transformer on them in the first place, Id be thinking about a proper line stage to drive them. Of course you could put any line stage in there you want, which blows the doors of creative choice wide open, but looking at the master section pcb's Ive seen photos of, it doesnt appear that the line stage would be very hard to build up assuming they dont use a crazy esoteric transistor that had the "sound".

anyway, just a thought.

dave

Were there originally direct outputs on the console to begin with, or are you just using an aux or something?

Click to expand...

The console did not have direct outs. Most people (myself included) are pulling the direct out off of the last coulpling cap before the bus section.

If you are just using the diret out as a monitor path, do you *need* to go through the trouble of balancing it with a transformer?

Click to expand...

The PM1ks are racked channel strips. From the unbalanced output I'm sending signals in parallel to a) the DAW input (which operates at +4 but most likely has an input impedance significantly higher than 600ohm, and b) my 600ohm console inputs for monitoring. I'm assuming that this is bad.

Why not take the unbalanced direct out you want to use and then feed that to a copy of the line driving stage that you can build on a little piece of perfboard inside the box? This way, you know you'll have the might to drive the transformer "properly" to whatever spec they had designed and you'll get the "sound" of the console in your direct outs.

Click to expand...

That's a damn good idea. For all those interested, here's the gain block in the master section:

http://img95.imageshack.us/img95/2267/pm1kamp7ko.jpg

-ben

Boy, it would so cool if we could come up with a nice line driving output stage and put together a group order on some small little pcbs and do a group order. I'm sure there would be enough interest to do it. Unfortunately I don't know enough to make it happen. Or do you think it would be something easy enough to breadboard and keep on breadboards? That might be a pain to do on 16 modules though.

Someone would have to be a guinea pig though! :green:

It would be my *GUESS* that the circuit would have a much easier time driving that balancing stage than it would the 10K+ input of your computer. You gotta figure it was designed to drive that gain stage in the first place, I would certainly try that and see where it gets you. Ive built more complicated stomp boxes on perfboard back in the day, you could probably put that circuit together in under two hours if you go real slow and tripple check along the way. Its helpful when you do a circuti like that on perf to print out the schematic and get a hilighter or a red pen or something and circle each component as you place it on the board and then highlight the "trace" as you solder the components together, makes it easy to keep track of the circuit that way.

The way I see it, if you are using the thing as a mini console, you might as well design it as a mini console, copying the line driver seems like the most logical thing to do. Unfortunately, thats where my good ideas end, hopefully one of the smarter guys around here can chime in on wether or not the spot you want to implement it will work, thats definitely outside the realm of what I do and dont know.

good luck

dave

Here's a little line driver that people are familiar with (Hamptone JFP module). I drew this up for the little 1.5-inch square perf-boards you can get at radio shack. Add 2 resistors and it will run off of 48vdc.

http://img328.imageshack.us/img328/7382/jfp1dp.jpg

I don't know if i can lay out the pm1k gain module on something that small. I count 22 passives, 1 diode, and 4 transistors. Furthermore I can't seem to find an equivalent to the 2sa841.

-ben