The No.1 Website for Pro Audio
API 550B racking with transformer input
Old 4 weeks ago
  #1
API 550B racking with transformer input

Hi everyone,
I'm racking a pair of API 550B (about 1990) and I realized they have unbalanced inputs and the SGN- is shorted to GND inside the PCB. So I decided to unbalance the incoming signal using a transformer (Carnhill VTB9071). Is is normal that, feeding the module with a 1.23Vrms signal (1kHz, measured form pin 10 to pin 8 on the card edge connector), I measure 1.46Vrms at output? It's like having a constant 1.5dB of gain...even with the EQ bypassed.
It seems to work properly but I have a few additional questions:
  • Why when I boost [email protected], the 1kHz sine wave is amplified by about 1.5 dB? How wide is the bell? I stop noticing gain when I set the boost freq at 180Hz
  • The two pos and neg signal at the module's differential output are different in amplitude. For example the pos is 0.9Vrms and the negative is 0.6Vrms with respect to Audio _GND, instead of having 0.75 and 0.75. I know that the important thing is the differential signal, but why it is asymetric?
  • Do I have to connect the Chassis_GND and where? To Audio_GND?

The modules are powered by a bipolar low noise power supply @ +/- 16V. Audio_GND is not connected to pin 3 of output XLRs, and not connected to input signal's GND too.

Thanks a lot for your support!
Regards
Stefano
Old 4 weeks ago
  #2
Gear Maniac
 
kludgeaudio's Avatar
 

Quote:
Originally Posted by tubeing View Post
Hi everyone,
I'm racking a pair of API 550B (about 1990) and I realized they have unbalanced inputs and the SGN- is shorted to GND inside the PCB. So I decided to unbalance the incoming signal using a transformer (Carnhill VTB9071). Is is normal that, feeding the module with a 1.23Vrms signal (1kHz, measured form pin 10 to pin 8 on the card edge connector), I measure 1.46Vrms at output? It's like having a constant 1.5dB of gain...even with the EQ bypassed.
That's a 10K 1:1 transformer. It needs to see a 10K load or it will ring. If the input impedance of the API doesn't match it, you will see ringing and overshoot.

Put a 1KHz square wave through it and look at the structure of the square wave. It should look nice and square... no ringing on leading and trailing edges, the corners should be clean and square and the top and bottom should be nice and parallel. If not... either change the transformer termination or change the transformer.

Be aware that it's a lot easier to make a good 1K:1K transformer than a good 10K:10K transformer, so unless you really really need that high input impedance you'd be likely better off with a lower ratio.

Every time you start seeing more gain than you expect with a transformer, suspect ringing. The scope will tell you everything you need to know... the meter only tells you some of it.

Every time you start seeing anything asymmetric, start worrying about second harmonic distortion. But don't worry about the values on the meter being asymmetric, look at the waveform.
--scott
Old 4 weeks ago
  #3
Lives for gear
 
nosebleedaudio's Avatar
 

Quote:
Originally Posted by tubeing View Post
Hi everyone,
I'm racking a pair of API 550B (about 1990) and I realized they have unbalanced inputs and the SGN- is shorted to GND inside the PCB. So I decided to unbalance the incoming signal using a transformer (Carnhill VTB9071). Is is normal that, feeding the module with a 1.23Vrms signal (1kHz, measured form pin 10 to pin 8 on the card edge connector), I measure 1.46Vrms at output? It's like having a constant 1.5dB of gain...even with the EQ bypassed.
It seems to work properly but I have a few additional questions:
  • Why when I boost [email protected], the 1kHz sine wave is amplified by about 1.5 dB? How wide is the bell? I stop noticing gain when I set the boost freq at 180Hz
  • The two pos and neg signal at the module's differential output are different in amplitude. For example the pos is 0.9Vrms and the negative is 0.6Vrms with respect to Audio _GND, instead of having 0.75 and 0.75. I know that the important thing is the differential signal, but why it is asymetric?
  • Do I have to connect the Chassis_GND and where? To Audio_GND?

The modules are powered by a bipolar low noise power supply @ +/- 16V. Audio_GND is not connected to pin 3 of output XLRs, and not connected to input signal's GND too.

Thanks a lot for your support!
Regards
Stefano
These are very wide at low boost, around 4 octave wide.
If both are the same In/Out levels I would not worry about the level increase..
10K:10K is a bridging transformer which works very good esp if they are a good one like Jensen.
Old 4 weeks ago
  #4
Quote:
Originally Posted by kludgeaudio View Post
That's a 10K 1:1 transformer. It needs to see a 10K load or it will ring. If the input impedance of the API doesn't match it, you will see ringing and overshoot.

Put a 1KHz square wave through it and look at the structure of the square wave. It should look nice and square... no ringing on leading and trailing edges, the corners should be clean and square and the top and bottom should be nice and parallel. If not... either change the transformer termination or change the transformer.
Thi API's input impedance is 18.5kOhm. And I connected the two secondaries in parallel to get a 2:1 transformer to avoid 6dB of gain only by using transformers. The recommended load impedance is 2.5k in this configuration...How could I match the impedance? Putting a resistor in parallel to the sec windings? This will lower the signal I suppose...

Quote:
Originally Posted by kludgeaudio View Post
Be aware that it's a lot easier to make a good 1K:1K transformer than a good 10K:10K transformer, so unless you really really need that high input impedance you'd be likely better off with a lower ratio.
I'm not a transformer expert, but Jensen, Lundahl and Sowter recommended a 10K : 10K for this application, so I followed their suggestions. I also have 600:600 transformers, but I don't know if they are suitable for this job...

Quote:
Originally Posted by kludgeaudio View Post
Every time you start seeing more gain than you expect with a transformer, suspect ringing. The scope will tell you everything you need to know... the meter only tells you some of it.

Every time you start seeing anything asymmetric, start worrying about second harmonic distortion. But don't worry about the values on the meter being asymmetric, look at the waveform.
--scott
I don't have an hardware waveform generator here, but I used the SW generator in Logic. With sines I didn't see any kind of distortion from 30Hz to 16kHz (using FFT on my scope), with square waves I see ringing, overshoot/undershoot even if I don't connect the input transformer, so I suspect that either the soundboard or the wave generation has problems...

Thank you very much for your reply!
Stefano
Old 4 weeks ago
  #5
Quote:
Originally Posted by nosebleedaudio View Post
These are very wide at low boost, around 4 octave wide.
If both are the same In/Out levels I would not worry about the level increase..
10K:10K is a bridging transformer which works very good esp if they are a good one like Jensen.
Thanks! I was only wondering why they should add gain even when bypassed...and if the input transformer could play some role in this.
In these days I'll try them connected to my audio chain to understand how they perform, while measuring their output at the same time to look for some strange behavior (if any).
About impedances, I fear I'll have to deal with some oscillations with no matching, and I really don't know how to address this potential issue. But we'll see in the next few days.
Regards,
Stefano
Old 4 weeks ago
  #6
Gear Maniac
 
kludgeaudio's Avatar
 

Quote:
Originally Posted by tubeing View Post
Thi API's input impedance is 18.5kOhm. And I connected the two secondaries in parallel to get a 2:1 transformer to avoid 6dB of gain only by using transformers. The recommended load impedance is 2.5k in this configuration...How could I match the impedance? Putting a resistor in parallel to the sec windings? This will lower the signal I suppose...
Yes, it will lower the signal to the point where it is unity gain. And the transformer will not ring.

Quote:
Originally Posted by tubeing View Post
I'm not a transformer expert, but Jensen, Lundahl and Sowter recommended a 10K : 10K for this application, so I followed their suggestions. I also have 600:600 transformers, but I don't know if they are suitable for this job...
In a perfect world this would be the case, but 10k:10k transformers are harder to make well than 600:600 transformers. With a 600:600 transformer you would have a 600 ohm termination resistor across the secondary (or 620 ohms since that's the nearest 2% value), which is good. However, anything you plugged into it would have to be able to drive a 600 ohm load, which might be a problem in a modern studio.

Quote:
Originally Posted by tubeing View Post
I don't have an hardware waveform generator here, but I used the SW generator in Logic. With sines I didn't see any kind of distortion from 30Hz to 16kHz (using FFT on my scope), with square waves I see ringing, overshoot/undershoot even if I don't connect the input transformer, so I suspect that either the soundboard or the wave generation has problems...
Take the API and transformer totally out of the circuit and see what the wave going into the API looks like.

This is a perfectly-terminated transformer: http://www.panix.com/~kludge/xformers/DSC00259.JPG
You can see the corners here are a little soft, because this is a kind of cheap transformer, but it's okay.

This is the same transformer with no termination resistor:
http://www.panix.com/~kludge/xformers/DSC00258.JPG

It's ringing like mad because of internal resonances in the windings.
Now, you measure this with a non-RMS meter and you're going to see a much higher voltage than you did before (which is as much an artifact of the meter as anything).

If you were to put too low a resistor across it, you would get this:
http://www.panix.com/~kludge/xformers/DSC00260.JPG

This transformer is over terminated and is losing high frequency response as a result.
--scott








Thank you very much for your reply!
Stefano[/QUOTE]
Old 4 weeks ago
  #7
Quote:
Originally Posted by kludgeaudio View Post
Yes, it will lower the signal to the point where it is unity gain. And the transformer will not ring.



In a perfect world this would be the case, but 10k:10k transformers are harder to make well than 600:600 transformers. With a 600:600 transformer you would have a 600 ohm termination resistor across the secondary (or 620 ohms since that's the nearest 2% value), which is good. However, anything you plugged into it would have to be able to drive a 600 ohm load, which might be a problem in a modern studio.
I could try to connect the transformer 1:1 instead of 2:1 as I'm doing now and put a resistor across the secondaries (in series) to lower the voltage to get half of the signal, looking if this is a better solution...or, simply, put a termination resistor across the two sec windings (now in parallel) to get the out of the API equal to the input while trying to lower the ringing...
With a potentiometer I could find the sweet spot in terms of ringing checking then if I introduced a gain or not.


Quote:
Originally Posted by kludgeaudio View Post
It's ringing like mad because of internal resonances in the windings.
Now, you measure this with a non-RMS meter and you're going to see a much higher voltage than you did before (which is as much an artifact of the meter as anything).

If you were to put too low a resistor across it, you would get this:
http://www.panix.com/~kludge/xformers/DSC00260.JPG

This transformer is over terminated and is losing high frequency response as a result.
--scott
I use a True RMS to measure and check everything with my scope to be sure. Thanks Scott for the suggestions, really appreciated! I'll try immediately...
Stefano
Topic:
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…
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump