diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Multi-Way (http://www.diyaudio.com/forums/multi-way/)
-   -   Re-Jigging the jig: Speaker testing device for Arta, Speaker Workshop & HOLMImpulse (http://www.diyaudio.com/forums/multi-way/118015-re-jigging-jig-speaker-testing-device-arta-speaker-workshop-holmimpulse.html)

MJL21193 23rd February 2008 06:21 AM

Re-Jigging the jig: Speaker testing device for Arta, Speaker Workshop & HOLMImpulse
 
2 Attachment(s)
About 3 years ago I started to seriously research speaker design and building. I had always built speakers, but that was by guess only. None of my "designs" had any science behind them.

I found Speaker Workshop, and decided I'd use that to help my new designs. I ordered the parts for a Wallin jig and set about building it. It took hours to build this and when it was done, it didn't work! I spent several more hours tracking down the cause of the problem.
Now, I'm in no way putting down Eric Wallin's design, but for what it does, it's too complicated. Trying to squish all of those switches and resistors and wires into that little project box is tough.

Originally, I only used SW for driver impedance testing, to determine the T/S parameters. Later, I decided I'd try to use it for frequency response testing, so I ordered the parts to build Wallins mic preamp. That was fine: I now had a preamp, a homemade mic and the test jig. I needed an amp. The only thing I had (at the time) was a vintage Pioneer SA-6800 - big and bulky.
This kind of setup in ok if you have a dedicated space, where you can leave everything connected and undisturbed. If you need to move things, then the setup doesn't seem so good.

With this in mind, about a year ago, I changed things. I rebuilt the jig on a circuit board and included the mic preamp. I then built an amp with an LM3886 chip amp. I packed all of this into an empty computer power supply housing. I had the power supply for the chip amp on a piece of plywood, sitting next to the new jig.
This setup was better, and I have been using it up till now. The only drawback is the separate power supply.

I thought it would be nice to have everything in one neat and tidy package. Something that was well designed and included all of the important features.

I started with the jig itself. I made some changes that, to me, made sense. I eliminated the 4/16 ohm calibration resistors and with it the switch that puts these in and out of the circuit. I also deleted the clamping zeners as these are a potential source for distortion.
I established a voltage divider for the amplified signal which will protect the sound card from the amp.
I made the switches more "friendly", with just 2 on the front panel for jig function: direct SW3 (puts the jig in loop back for sound card calibration) and frequency response/impedance measure SW1.
The power switch for the amp (SW2) also cuts the line out from line in, allowing all of the connection cables between the jig and the sound card to remain in place.

gootee 23rd February 2008 06:29 AM

That sounds like a good little unit, John! I wish I had one. Do you have any photos?

pinkmouse 23rd February 2008 09:23 AM

2 Attachment(s)
I'm building something not dissimilar, only I'm keeping the mic pre separate.

MJL21193 23rd February 2008 01:02 PM

2 Attachment(s)
Quote:

Originally posted by gootee
That sounds like a good little unit, John! I wish I had one. Do you have any photos?

Hi Tom,
Yes, I've got some pictures of my progress so far, and I'll be posting those as I go along. As it is not fully complete, I'd appreciate any comments for improvement, especially with the jig circuit above.

I wanted to make the amp for frequency response testing as compact and simple as possible. I used the same circuit design I've used for other chip amps I've built, but I reduced the supply voltage to +/-18 V and reduced the gain to 11 V/V. Typically, this amp will only supply a watt or two in most instances.

By keeping the supply voltage low, I could use smaller 25 V electrolytic, making the board layout as compact as possible. Making the board double sided, I created pads on the top copper and on the bottom copper to surface mount the leads of the LM3886, rather than thru hole mounting. This allows for wider traces in a more compact layout.

Here's the schematic for the amp:

MJL21193 23rd February 2008 01:27 PM

2 Attachment(s)
Quote:

Originally posted by pinkmouse
I'm building something not dissimilar, only I'm keeping the mic pre separate.
That is very professional looking Al. How much more work do you need to do on it? What did you use for the front panel labeling?
On mine, I tried to use the laser print transfer method, but it kind of smeared (I'll post a pic later).

For the mic preamp, I used Eric Wallin's design, but made some changes. Having built the preamp before, I know how well it works - low noise and efficient. A 9 volt battery lasts a long time.

I used the NE5532 for the opamp. This opamp is as good as they get IMO.
No major changes to the circuit, other than eliminating the clipping indicator. Test that I made during the construction of this preamp show clipping to be well above the practical input level.

In the original design, the gain is selectable between 2 and 20. I found the low gain useless - I never used it as the output was not enough. Likewise, the high gain seems just adequate sometimes so I went ahead and changed that, increasing the gain to ~7 and 27 for low and high.

pinkmouse 23rd February 2008 02:04 PM

John, it is indeed toner transfer. The secret with front panels is to scuff them up a bit so the toner adheres. I just ran an orbital sander loaded with 150 grit over it lightly 'till I got an even look. As for completion, I'm working on it as we speak, just waiting for the chipamp board to etch at the moment. I hope to be finished by close of play today, then some testing tomorrow. ;)

My mic pre is the kit that Vikash sells, AFAIK, it's not disimilar to the Wallin one

MJL21193 23rd February 2008 03:10 PM

2 Attachment(s)
Quote:

Originally posted by pinkmouse
John, it is indeed toner transfer.

Hi Al,
My trouble started when I tried to print the labels in a mirror image. For some stupid reason, my printer (HP 3050) software doesn't support it, so I had to improvise. I used MS Word to type the words on the screen, then did a screen capture to the clipboard. Pasted this to Paint and reverse and resized the type.
A genuine pain in the ****.
There must be a better way.
Then, the front panel on mine is solid white oak. I sanded this as smooth as I could get it, but it still has open grain.
I can't say I'm thrilled with the results, but it's good enough for what it is and as long as I can read it, it'll be fine.

Here's a pic of the final finished product. Overall dimensions are 3" high, 5 1/2" wide and 7 1/2" deep. Jacks for sound card connection, mic input, power mains input and a switch for mic gain are on the back (more pics to come).
This morning I added a "power on" LED for the mic preamp. This will remind me to turn it off :)

daatkins 23rd February 2008 05:50 PM

2 Attachment(s)
Haven't we all been busy!

The attached pic is my take on the all-in-one Speaker Workshop jig. It contains an L165V chip amp, my own design mic pre-amp and all the signal switching of the Wallin jig but compressed onto two rotary switches. I've also added a spectrum analyser function for measuring the frequency response of active crossovers and the like.

I'm currently in the process of redesigning the jig - squeezing it into such a small case meant some compromise needed to be made. Swapping the ghastly L165V power op-amp for a LM3886 is on the agenda as is improving signal routing.

Nice one,
David.

daatkins 23rd February 2008 05:53 PM

2 Attachment(s)
And the schematic...

gootee 23rd February 2008 06:04 PM

Quote:

Originally posted by MJL21193



Hi Tom,
Yes, I've got some pictures of my progress so far, and I'll be posting those as I go along. As it is not fully complete, I'd appreciate any comments for improvement, especially with the jig circuit above.

I wanted to make the amp for frequency response testing as compact and simple as possible. I used the same circuit design I've used for other chip amps I've built, but I reduced the supply voltage to +/-18 V and reduced the gain to 11 V/V. Typically, this amp will only supply a watt or two in most instances.

By keeping the supply voltage low, I could use smaller 25 V electrolytic, making the board layout as compact as possible. Making the board double sided, I created pads on the top copper and on the bottom copper to surface mount the leads of the LM3886, rather than thru hole mounting. This allows for wider traces in a more compact layout.

Here's the schematic for the amp:

This all sounds good.

One thought that just occurred to me: A higher chipamp gain than 11 V/V might give better performance. If so, you could just put a resistive divider before the input, and keep the overall gain wherever you want it.


All times are GMT. The time now is 08:29 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2