The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Vint 600Ohm to Modern matching Restr 500 Series Preamps
Old 2nd September 2007
  #1
Gear Nut
 

Vint 600Ohm to Modern matching Restr

Howdy folks,

Can some one(s) lay out the exact (as much "detail for dummys" as possible please) way to assemble the 600 (620) ohm resistors in laymans terms (exactly where they connect to).

I have a 600ohm (transformer line stage final) and want to run it into Modern (10k?) gear. I was told to put 600ohm 'matching resistors' on the output of the unit (mic pre).

It has TRS/TS outs, so i was gonna wire the TRS's to a Terminal strip and configure it there.

QUESTION is, where EXACTLY do the Res's go (as far as connections, see below) ?

- Do i need 2 per channel?

Do they go like this:
-a resistor between Hot and Ground
-another between Cold and Ground

OR:

-between Hot and Cold?


Also, does this really make a sonic difference, if so-how and what?

....do tell

Thanks much,
BZ

Last edited by billzoe; 2nd September 2007 at 07:45 PM.. Reason: mispell
Old 2nd September 2007
  #2
Lives for gear
 
brianroth's Avatar
 

Most output transformers require a "proper" load across the secondary winding. Often, 600 Ohms is the proper value, but sometimes a somewhat higher loading will result in better performance,

The loading resistor goes across pins 2 and 3 of an XLR connector...NOT from one lead or the other to ground.

I suggest a 1/2 Watt resistor, and the exact value is non-critical. A 620 Ohm carbon (standard value in the 5% series) will work as well as a 1% metal film.

Bri
Old 2nd September 2007
  #3
Gear Nut
 

Thanks Brian,

Pins 2 and 3 on XLR are (+) and (-) right ?

.....its trs , so i am converting , i think Tip is (+) and Ring is (-) , Sleeve is Grnd...

...so just (1) Res from Tip TO Ring?

and NOTHING goes to ground?

Sorry for the dumm quest's.

Thanks MUCH!!!

BZ
Old 2nd September 2007
  #4
Lives for gear
 

Hi
Yes, tip to ring.
If a transformer is not loaded with the designed impedance then it is liable to 'ring' which will usually appear as a significant 'boost' in level at some frequency determined by the transformer and the cable capacitance. This may be in the audio band or up to 200KHz, maybe more.
This boost may be attractive or it may be catastrophically bad for 'your sound', it is up to you.
Most designs of equipment will aim to have the response 'correct' for (say) 600 ohm load and a practical length of cable (say 20 Metres).
Matt S
Old 4th September 2007
  #5
Gear Nut
 

Matt,

Thank you..... that is quite helpful. I'll give it a go.. I think i noticed what you were describing , it was a flubby sound sometimes , or ringy , but seemed to move around the spectrum depending on source music.

Thanks again all!!
Old 4th September 2007
  #6
Motown legend
 
Bob Olhsson's Avatar
 

Some of us anal perfectionists use two 1200 Ohm resistors in parallel.
Old 8th September 2007
  #7
Lives for gear
 
emrr's Avatar
I like 680's for the simple reason that there is 5K input gear out there and the combination works out to be a 598 ohm load to previous unit.

680 into a 10K = 637 ohm load to previous unit.

620 into a 10K = 584 ohm load to previous unit.

620 into a 5K = 552 ohm load to previous unit.

600 into a 10K = 566 ohm load to previous unit.


Hope I got all that right.
Old 8th September 2007
  #8
Lives for gear
 

Hi
There is no need to be too accurate as resistor tolerances used at the time were usually 10 percent. Since 1% types are easy and cheap go for them but the exact value is not so important as replugging to a different unit will make your calculations irelevant. The important bit is that it is not 10K or more and that it is mostly resistive.
620 or 680 are both fine.
Matt S
Old 9th September 2007
  #9
Gear Nut
 

Thanks much all!!

can a patient soul attempt to help me understand WHAT we are doing here with this Resistor, for example:

-If we are runining into 10k (or whatever) , we want to reduce that , right?

-So, is it that we are halving or dividing the 10k up by putting a 600 Ohm across terminal?

i dont know electronics alot, but i seem to remember something about Series or Paralllel either halves or doubles RES - is that what we are doing, if so , which is which?

Sorry for idiot question, perhaps if i can understand and figure out how to calculate it with diff Gear i'm running into, it would be helpful.

Cheers,
BZ
Old 9th September 2007
  #10
Lives for gear
 

Hi
In an attempt to make it simple I will not put precise values.
Basically transformers need to be loaded with a relatively constant load resistance as their impedance is complicated and varies with frequency. When presented with a capacitance (the cable) it will form a 'tuned circuit' which will have a resonant frequency.
In order to be useful all circuits involving transformers will state a value of resistance which will ensure that the tuned frequency is 'inhibited' such that the overall response is as flat as possible. To make life 'simpler' it was agreed many years ago that 600 ohms would be used as a standard, I believe that this came about from telephone circuits where it was found that LONG wire circuits (miles) tended to have an impedance of about 600 ohms.
In the days when loss of POWER was important the source and load should be matched, so the 600 ohm standard was upheld.
Nowadays with amplification being cheap we use a power inefficient technique of low output impedance and high input impedance (which does have other benefits).
The result of this is that most modern gear has an input impedance of 10K minimum, often more.
Hanging this on a transformer output intended to drive 600 ohms will lead to insuficient 'damping' of the transformer / wire tuned circuit, hence it will exhibit peaks.
Simply stuffing a resistor of a little over 600 ohms across the output will provide the correct damping, thus correcting the response. In theory you should check the input impedance and then add the correct loading but life is a little short, besides, the response may not have been so flat anyway even with 600 ohms as it will depend on the cable capacitance. Thus the addition of either 680 or 620 ohms across the output is a reasonable compromise unless you have the gear and patience to check the system fully. This also assumes you really want it flat anyway!!
Matt S
Old 9th September 2007
  #11
Lives for gear
 
emrr's Avatar
Don't get hung up on the precise values of my examples other than to recognize the fact that an exact 600 ohm match can turn out to be too low, depending on the piece of gear, and once you get up to 680 or so you are generally in the safe zone of being either exact or a little high in most cases. In many cases a 750 or 1000 ohm load resistor would be sufficient.

Standard application of ohm's law:

680 (load resistor) x 10,000 (theoretical standard input impedance) = 6,800,000.

6,800,000 divided by 10,680 (load resistor + input impedance) = 636.7 ohm load presented to previous piece of gear.
Old 10th September 2007
  #12
Lives for gear
 
ulysses's Avatar
To put it more plainly and in terms you can use in the future:

Put two resistors in series, and the resulting resistance is the sum of those two resistors:

10K + 620 = 10620 ohms.

But that's not what we're doing. We're putting them in parallel.

Put two resistors in parallel, and the resulting resistance is the Reciprocal of the sum of the reciprocals. When the two resistors are equal, the calculates out to simply being half the value of each individual resistor. When they're not equal, then you need to do the math:

R = 1/[(1/R1)+(1/R2)]

So, for the 10k load with a 600 ohm resistor strapped across it, you have:

R = 1/[(1/10000)+(1/600)] = 1/(0.0001+0.001666) = 1/0.001766 = 566 ohms. Even if you're bad at math, this is a simple calculation and it's really easy to do with a calculator. Especially if the calculator has a "1/x" key. Nobody should be allowed to own recording gear without also owning a decent calculator. I use the "1/x" key numerous times every single day. You have to be able to do some basic calculations if you want to reliably know what you're going to get with different pieces of gear connected together.
Old 10th September 2007
  #13
Gear Nut
 

Matt , Doug, and Justin - THANKS MUCH !!!!!


This is exactly what i was trying to find out and very well written and elaborated, thanks Justin for Dummy (me) assistance!

Matt, one more thing, i know it is correct, but my thick head seems stumped at why it is that 10k is not enough Dampening, isnt 10k MORE resistance than 600 ? Somehow more is less ??

gotta a brain ache....arggh !!

I know you guys are laughing, but i never did pass that Electronics 101 course



Cheers,
BZ
Old 10th September 2007
  #14
Gear Nut
 

"This also assumes you really want it flat anyway!! "
Matt S




.....right, yes i dont want it Sterile, i guess i just want it Stable, but still Colored [coloured-i lived in UK once ] (its Passive Inductor EQs, Mic Pres) - so i guess that means just get "CLOSER" to 600 ohms than 10k, accuracy not extremely important, eh?

Makes one appreciate the skill of the actual on-site engineers!!

thanks,
BZ
Old 10th September 2007
  #15
Lives for gear
 

Hi
The inductance (transformer) capacitance (cable) and resistor are in parallel and the lower the resistance the more 'power' is wasted as heat from the tuned circuit.
A perfect indictor and capacitor in parallel would lose no energy and would oscillate forever once started with an impulse of electricity. Perfect caps and inductors don't exist in the real world!
Matt S
Old 10th September 2007
  #16
Lives for gear
 
A27Hull's Avatar
 

+1 for discussion,
Old 10th September 2007
  #17
Lives for gear
 
A27Hull's Avatar
 

Quote:
Originally Posted by ulysses View Post
So, for the 10k load with a 600 ohm resistor strapped across it, you have:

R = 1/[(1/10000)+(1/600)] = 1/(0.0001+0.001666) = 1/0.001766 = 566 ohms.
Going further, the sending unit (output transformer) is seeing a 566 ohm load, give or take but wants to see 600 ohms.

Is this right?

If you wanted the full 600 ohm load to be seen, you'd need a higher value resistor across the output, perhaps a 638.5 ohm resistor.

But I don't thinks that is a practical value...is the closest 640 ohms?

[1/Rt=1/10K + 1/638.5 = approx. 600.18]

[1/Rt=1/10K + 1/640 = approx. 601.50]
Old 10th September 2007
  #18
Lives for gear
 
ulysses's Avatar
The point is that certain transformers might have some high-frequency ringing if it's very lightly loaded (like with 10k or even 100k). By getting the loading *near* 600 ohms, that ringing is tamed. 600 ohms was the nominal standard at the time, but any particular transformer model may have differed slightly (or greatly). In fact, some transformers are designed to work best when loaded very lightly (in other words, connected to a high impedance load). If you wanted to know the *exact* ideal load impedance, you would put a square wave through the device in question, and look at the output on an oscilloscope. You would adjust the loading until the output looked as much as possible like the input. In some cases, a simple resistor will not achieve this. In any case, calculating the exact correct value resistor to give you exactly a 600 ohm load is pointless unless you first examine the output you're terminating to find out what its ideal load really is. As a practical matter, getting within a range of about 50-200% of the ideal value will probably get the job done to within your ability to ascertain any difference. Remember, you are going for good sound after all. If you can't hear or measure a difference between a 600 ohm load and a 620 ohm load, than what does it matter? (Unless you hear or measure in a way that misses what somebody else is going to hear (or measure) on your recordings). If you use your ears and/or an oscilloscope, you'll probably find that a 1k resistor would be just fine. It's not that big of a deal.
Another thing that's been alluded to here is that in some cases the unloaded transformer, with its HF ringing, is actually desirable. The most obvious example is the Neve 1073, which many people removed from consoles and racked up without the 600-ohm load that the console provided. The HF ringing is actually a part of what many people know as The Neve Sound, even though it was not originally intended and was not present on the original recordings that made That Sound famous.

You responded well to the formula for calculating parallel resistance, so let me lay a couple more, far more imporant formulas on you that explain why a lower resistance is a bigger load:

Ohm's law:

E = I * R

Voltage (Electromotive force) equals Current (I) times resistance (R). Can also be written I = E/R or R = E/I

So, for a given signal voltage, a lower load resistance will cause more current to flow.

P = E * I

Power (P) equals voltage times current. So, for a given signal voltage, a lower load resistance will cause more current to flow, which means that more power is delivered to the load.

Can also be written P = E^2/R

These two formulas are the foundation on which all electronics theory is based.
Old 10th September 2007
  #19
Lives for gear
 
A27Hull's Avatar
 

Thank you very much Ulysses!

Brilliant!
Old 10th September 2007
  #20
Lives for gear
 

Hi
Thanks Ulysses, I don't do maths before breakfast (or after really)!
I remember years ago 'tuning' the desks in a radio station where for IBA code of practice you had to go through the newsroom desk, main studio desk, Continuity desk then main racks to be presented at the lines for the transmitter.
The transformers used generally had a peak at about 35KHz and I spent half the night finding the corrct loading (res + cap) to get 4 pieces of equipment and hundreds of metres of cabling 'flat' to 25K then gradually falling above this.
The following morning my boss came in and took them out, until it was observed that the system now 'failed' code. My components were then reinstated and happiness reigned.
Many of the early Audix desks used transformers (like the Neve's of the period) and we had tables to show what loading was needed for each transformer type in each situation, often involving resistance (or two) and a capacitance.
Incidentally similar reasoning is involved with 50 / 75 ohm cabling for video and SPDIF applications, in that the correct termination is necessary to ensure the system works as intended (prevention of reflections and level matching).
Matt S
Old 10th September 2007
  #21
Gear Nut
 

"Another thing that's been alluded to here is that in some cases the unloaded transformer, with its HF ringing, is actually desirable. The most obvious example is the Neve 1073, which many people removed from consoles and racked up without the 600-ohm load that the console provided. The HF ringing is actually a part of what many people know as The Neve Sound, even though it was not originally intended and was not present on the original recordings that made That Sound famous."




Again, great help from all, thanks!

The above quote is interesting , so this means "a LITTLE ring is a good thing" - wow !!!

I totally would've overlooked that - being most of the time i DO want the Colour, if not i'd try to be more exact (now) or just use different units.

Cheers all!
BZ
Old 11th September 2007
  #22
Lives for gear
 

Hi
The 'ring' is like having a peaking EQ with say 4 - 6 dB boost but as it is probably centred at say 30KHz it is not so obvious but just adds 'sparkle'.
The transformers on my units have a peak at around 80KHz unless damped.
Matt S
Old 11th September 2007
  #23
Gear Nut
 

Matt,

thank you for the above, it helps to see it in a simple way.


i seem to notice also a fleeting (elusive) or intermitent Low Frequency oddity when Transformer is not loaded, almost like 2 woofers out of phase.

Its hard to verify because it doesnt do it all the time and oft if you try to replicate it, its not there - seems source and cable capacitance (?) dependent, but i am guessing.

is this in my head or is it another or related phenomena to the Ring ?

Cheers
BZ
Old 11th September 2007
  #24
Lives for gear
 
nosebleedaudio's Avatar
 

It also depends on the transformer, Jensens have very little peaking and dont have to be loaded with a 600 ohm load.
The output circuit wants to see a load as well, before the output transformer.
And loading the secondary reflects that load back to the output circuit.
Old 11th September 2007
  #25
Lives for gear
 

Hi
Yes some transformers can exhibit very little ringing but this is due to design and with well defined input and output conditions.
As Nosebleed was suggesting we have only started the subject really as overall response depends on many other factors such as the impedance of the amplifier driving the transformer, load capacitance and whreter there is a lot of feedback used on the drive amplifier.
Suffice to say for now, unless you have a pile of test gear available stick a resistor of about 680 ohms across the output and you are at least in with a chance.
Matt S
Old 12th September 2007
  #26
Lives for gear
 
ulysses's Avatar
Unless the particular piece of gear you're terminating can't drive a 600 ohm load. In that case, don't.
Old 17th September 2007
  #27
Gear Maniac
 

I have an API 512C module mounted in a 6 slot lunchbox frame. I've been plugging it straight into a 002 line input with a stock balanced cable. The 512C module has an output impedance of less than 75 ohms. The 002 has an input impedance of 10 meg ohms (?) I think. I've always noticed that the tone is not exactly smooth, but very bold in the upper mids and highs...

If the 512C is mounted in a full blown API console, is it loaded with 600 ohms? I noticed solder pads on the lunchbox PCB labeled 'output load 1/5K' so maybe I'm onto something here?
Old 17th September 2007
  #28
Lives for gear
 

Hi
Sorry I don't know the specifics of the original application of your modules but in general all transformer outputs should be 'terminated' with a resistance within a certain 'range' of values.
Most op amps would drive 600 ohms but may well be happier with 1 - 2 K ohms (don't generally need termination) but as you have a transformer there should be something so it is probably worthwhile getting yourself a handful of resistors say 1K, 2K and give them a try out.
Matt S
Old 17th September 2007
  #29
Lives for gear
 
ulysses's Avatar
No, not all transformers. There are plenty that are designed to expect an unterminated, or "bridging", load. Pretty much anything designed in the last 3 decades or so by people who know what they're doing (like Jensen, for example) knows that it's living in a voltage transfer world and that it is unlikely and undesirable that it be terminated with anything like a 600 ohm load. Consequently, good modern line output transformers are designed to have good frequency response without such a termination.

Looking at data sheets for good Jensen output transformers, you can see that their test methodology does include termination. However, I believe this is in line with the fairly common practice of designing gear that is capable of driving a load, but that also performs well unterminated, since the load is not expected in practice.

At the other end of the signal chain, there are ribbon microphone output transformers. There's been a lot of excitement about mike preamps that allow you to change the load presented to this transformer, based on the idea that ribbon mikes universally "like" to be loaded and sound better that way. But RCA in particular designed output transformers that were intended to be used unloaded, and which sound best that way. So you can see how the ideal load impedance on the secondary of any transformer is, at least to some extent, a question of design choice. There was a time when a 600 ohm load was a standard and equipment was designed to fit that standard. 600 ohms is no longer standard, and though there is not a new standard that has completely replaced it, there is at least a common practice of loading low impedance sources with high(er) impedance loads. That is how line output transformers should be designed today.
Old 17th September 2007
  #30
Lives for gear
 

Hi
A load of some sort is expected and 'purely' capacitive would produce uneven response unless tightly specified. Certainly not tolerating cables of a couple of feet to several hundred feet.
The original post was about 'vintage' gear .
Matt S
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Matt Hepworth / So much gear, so little time
35
Giganova / Low End Theory
4
Honest Abe / Gearslutz Secondhand Gear Classifieds
11
clipgod / Gearslutz Secondhand Gear Classifieds
0

Forum Jump
Forum Jump