TZ81 Transistor

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
TZ81 was a small-signal NPN made by Sprague. I can't find any other data, but some cloners of vintage synths report successful substitution with the very common 2N3904.

Are you building a 1588B clone? That's an unusual choice.
 
Are you building a 1588B clone? That's an unusual choice.

No, but I was kicking around the idea of trying a few different parts substitutions on some existing ones. I can't imagine that those tiny little input transformers help the sound of these things, and I thought maybe some different transistors and caps might lead to some tonal differences.
Any suggestions on things to try?

I bought an Altec 1592A Mixer for $30 a couple years back on Ebay, and it came with 3 1588B modules. I haven't really been using it too much (sounds interesting on electric guitars).
 
I was given a 1592B and a 1599B extender loaded with 1588Bs (and a few of the now-coveted "red" line transformers) about 12 years ago. I still haven't fired them up!

The 1588B uses a fairly classic 3-transistor "ring of three" variation that can be (and has been) adapted to a wide variety of uses. I imagine a version using 2N3904, 2N3906 and something fairly beefy for the output emitter follower (something with max ratings like the 2N4401) could make for some cheap fun.
 
Followup: a quick check on the Lying-Assed Simulator(tm) indicates the circuit as-drawn should work fine with 2N3904, 2N3906 and 2N4401. To net a few more decibels of headroom, increase Vcc to +24V and use a 1-watt resistor for R7.

The transistors will be operating within their ratings--but if Q3 (which is dissipating over 100mW) gets noticeably warm, add a clip-on heatsink. Max output at 1% distortion is about +17dBM.

If the input transformer is truly tiny, that's likely the main limiting factor. So don't expect huge improvements in the sound simply from swapping transistors.

Hey, while you've got your 1588B open, will you take some pics?
 
I just pulled one my 1588Bs out of its can the other day. Then today I found this in the description of an evilbay auction. I wanted to throw it out there for discussion.

The seller's evilbay name is nclayton. I've actually bought some transformers from him. I'm going to send him an email requesting more info on this:

"The 1592B is a modular mixer designed to take up to five inputs from a variety of types of sources using dedicated plug-ins for the particular source to be used. Usually they were equipt with 1588 series microphone preamps to be a 5 into 1 mic mxer. They're very good sounding units -- they use Peerless transformers on every input and the output; they use all class A circuitry with all discrete transistors, no opamps; and they have a nice two band tone control with gentle broad boosts and cuts, which can actually be surprisingly usable, subtle and nice for general brightening or warming (it can also be switched out of the circuit completely). I'm a big fan of the 1592 sound, and have probably owned ten of them in all over
the years (both A's and B's). With the 1588B input module their topology is suprisingly similar to a Neve 1073: transformer in -> three cascaded transistor single ended preamp with feedback controlled gain (through the pad control)-> out to volume control and EQ -> three stage output amplifier directly coupled with the output transistors pulling their DC supply
current through the output transformer. Their biggest drawback is that when used for recording in a modern studio, they're basically just a single channel preamp since there's only one output. Every now and then I
find myself using a second channel either to mix two mics directly to tape or as a second input not to be used at the same time as the first (if that makes sense), but on the whole it seems like a waste that you can't use more of the inputs at once. I've bought and sold a few of these on ebay, and whenever I have, the question always comes up "can these be modified to have direct outputs" to which the answer is a
qualified "no".

The problem is that each 1588 module provides only about 33 dB of gain, then they're mixed together and the output amp adds the additional gain needed to bring the mic signal up to line (+30dBm max). The mixer as a whole system has TONS of gain, 87dB worth of gain, but as you can see, direct outputs taken from the 1588's would be pretty low level, good for very loud sources like drums or cranked up guitars, or maybe medium level sources if your recorder uses -10 inputs, but unusable for quiet sources, low output mics, etc.

So....Here's how I'm modifying all my 1592's now. The first three preamps (the ones with controls left of the VU meter) are stock, mixing into the master volume and output amplifer as usual. The last two (right of the VU meter) are rewired for direct output.

IN ADDITION, the two 1588B preamps in the direct output slots have been internally modified to increase their maximum gain. These 1588B's are modified for about 52dB, much more than the stock 33dB -- still on the low side for a mic pre, but much much more usable. 1588C's are not amenable to this type of gain mod, so I only use modified B's in these slots. Most 1588B's are getting old and a bit noisy, so while I have the cans opened up I also replace one critical carbon comp resistor with a metal film, and one electrolytic capacitor.

The extra gain is a result of decreasing the 1588B's internal negative feedback, kind of what the pad control does but in reverse. The 1588B circuit, open loop without any feedback, is capable of much more gain than they actually use. They were intentionally brought low by Altec so that when the mixer was used with 1588's for mic and 15095 transformers for line inputs, the mic inputs would have 45dB more gain than the lines (since the 15095 stepped down gain by about 12dB). The low gain of the 1588 modules was strictly a matter of not overwhelming the whole system with much more gain than was necessary. In fact, they're capable of much more gain even with more than sufficient linearizing feedback still applied. So, *theoretically* this modification increases distortion, but in practice the biggest difference in sound between these and normal inputs is that these don't go through the output amp, and the output amp does add some coloration. To make a long story short, the direct preamps don't sound exactly like the mixed preamps, but they do sound cool, and most importantly, with this 1592 you're getting three simultaneously usable mic channels instead of just one, and two of which can be used for stereo.

The outputs of the 1588B's are, of course, unbalanced. On this mixer I've added a pair of transformers to balance the outputs. I chose to use Peerless 4834 transformers for this job for a few reasons. First of all, they're about the same age as the mixer, and being original Altec gear, hey're a good "cosmetic" match I guess you could say (the 1588B's use the
Peerless 4838 input transformer, for example). Second, they're a one to one ratio transformer, so they don't loose any gain and have tremendous bandwidth and accuracy. Thirdly, they're NOT 600:600, they're actually rated for somewhat higher impedance. This is important since the way this circuit is laid out, the transformers come directly after the fader pots with
no buffer in between. In this circuit 600 ohm transformers could noticeably loose some bass due to the source impedance of the pots, but the 4834's have high enough primary inductance to keep a nice flat and rich bass response all the way to the bottom no matter where the fader pots are set. This is all worth mentioning, because these transformers DO have a downside which is that for best performance the direct outputs shouldn't be plugged into anything with lower than about 2K impedance, and preferrably higher. I doubt if this will be a problem, since most recorders/soundcards are 10K or higher."
 
[quote author="rafafredd"]Is there a schematic for those little things on line?[/quote]

From the earlier posts, it sounds like NY Dave has one. I can't find it searching the board or the web. I guess I'll trace it out, I need the practice. Though it would be nice to have Dave's to check against.

Dave can we have your schemo? Pleeeeease. :green:
 
[quote author="NewYorkDave"]Thanks to Scott Dorsey, who faxed these to me originally.[/quote]

Knowledgeable, generous, and humble to boot. What a guy!

Thanks Dave. :thumb:
 
[quote author="NewYorkDave"]The 1588B uses a fairly classic 3-transistor "ring of three" variation that can be (and has been) adapted to a wide variety of uses.[/quote]
I recall someone (you ?) posted a scan of such a "ring-of-three" circuit.
Can't find it back, do you recall it or still have it somewhere ?(IIRIC it was with tubes)

Thanks !

Peter
 
Since Peter revived this thread, I'll pose a question.

Why not use a pot as a variable resistor hung on the PAD pin for a continuously variable gain control?

See Altec Tech letter for details on controlling the gain of the 1588B:
http://alteclansingunofficial.nlenet.net/publications/techletters/TL_206.pdf
 
[quote author="skipwave"]Since Peter revived this thread, I'll pose a question.

Why not use a pot as a variable resistor hung on the PAD pin for a continuously variable gain control?[/quote]

W.r.t. gain-control, it'll need to be a revlog-pot, but that's not too big a problem.
I'll likely go for a rotary switch, feels nice and gives a bit more precise steps. Can be a nice combination with the 1k output-pot.

Bye,

Peter
 
I can't find the "ring of three" example I posted originally, so I've drawn a new one just now.

RingOfThree.png


This is the basic concept; there are many variations. Look closely and you'll find the ring-of-three lurking inside many of your favorite preamps, etc.
 
[quote author="NewYorkDave"]I can't find the "ring of three" example I posted originally, so I've drawn a new one just now.

RingOfThree.png


This is the basic concept; there are many variations. Look closely and you'll find the ring-of-three lurking inside many of your favorite preamps, etc.[/quote]
Thanks ! That's the one.

Nice that there are most of the time several ways to look at one & the same circuit :thumb:
Like this one you posted on the previous page: http://electronicdave.myhosting.net/miscimages/1588B-3.gif
'a ring of three' (indeed).
Or call it for instance a Sziklai-pair (or CFP) with an inserted emitter follower.
That EF solves the limitation of the CFP that you can't get lots of gain from it in practice. With just that one additional active device that's solved.

Cheers,

Peter
 
And if you can build one just like the drawing, I'll buy you a beer 'cause it would be tricky to get all the DC operating points just right :green:

(Before someone asks: no, that's not a DIY project. It's just an outline).
 
[quote author="NewYorkDave"]And if you can build one just like the drawing, I'll buy you a beer 'cause it would be tricky to get all the DC operating points just right :green:

(Before someone asks: no, that's not a DIY project. It's just an outline).[/quote]
I better don't even try (and won't Google to copy a dimensioning from somewhere else either :wink: )

I figure that making the feedback AC-only and giving the first triode some stuff for it's own DC operating point could 'decouple' the dimensioning-task somewhat. Perhaps the same for the second triode as well.
But that would be cheating :sam:

Bye,

Peter
 
[quote author="skipwave"]Since Peter revived this thread, I'll pose a question.

Why not use a pot as a variable resistor hung on the PAD pin for a continuously variable gain control?
[/quote]

I'm afraid the 1588B-for-sale made me revive more than just one Altec 1588B thread :cool: :wink:


For good measure, here's some more recent tech-talk going on:

Altec mic pre

Altec 1588B

(if there's enough traffic & interest we might try to continue in one 1588B-thread).

Bye,

Peter
 

Latest posts

Back
Top