QUAD 707 - DIY

This is an amplifier I made a long time ago
But due to the parts at that time not meeting the specifications. I made it around 2012.
But the test results were very poor. So I have been improving it.
I have recently obtained some very good transistors. The test results in the new version make me feel very good.
It has almost the same performance indicators as the machines produced by QUAD's original factory.
So I'm going to take the time to assemble it and listen to it myself. I will update the assembly steps.
of course. I am also an amateur businessman. I will provide it to others. But the condition is.
This product needs my approval. Otherwise, I will throw it away. For example, the QUAD707-2012 version. I have never had anyone sell it before.
 

Attachments

  • QUAD707-1.jpg
    QUAD707-1.jpg
    321.7 KB · Views: 636
  • QUAD707-2.jpg
    QUAD707-2.jpg
    392 KB · Views: 589
  • QUAD707-3.jpg
    QUAD707-3.jpg
    279.9 KB · Views: 595
  • QUAD707-4.JPG
    QUAD707-4.JPG
    544.7 KB · Views: 497
  • QUAD707-5.jpg
    QUAD707-5.jpg
    115.3 KB · Views: 450
  • QUAD707-6.jpg
    QUAD707-6.jpg
    102.7 KB · Views: 386
  • QUAD707-7.jpg
    QUAD707-7.jpg
    152.2 KB · Views: 371
  • QUAD707-8.jpg
    QUAD707-8.jpg
    144.7 KB · Views: 356
  • QUAD707-9.jpg
    QUAD707-9.jpg
    119.5 KB · Views: 460
  • 散热器尺寸图.jpg
    散热器尺寸图.jpg
    75.6 KB · Views: 467
  • 散热器尺寸图2.jpg
    散热器尺寸图2.jpg
    53.8 KB · Views: 598
  • Like
Reactions: 1 user
This is the chassis I purchased on Taobao
About 38 dollars
I use software to draw a fixed installation hole for the power tube.
I don't know how to make holes myself. That screw hole is too troublesome.
I will have the chassis manufacturer drill the screw holes.
 

Attachments

  • 机箱.jpg
    机箱.jpg
    61.5 KB · Views: 210
  • 散热器尺寸图.jpg
    散热器尺寸图.jpg
    75.6 KB · Views: 210
  • 散热器尺寸图2.jpg
    散热器尺寸图2.jpg
    53.8 KB · Views: 205
Chassis punch will easily make holes larger than 12mm (0.5 inch) after drilling a small hole.
https://www.harborfreight.com/knockout-punch-kit-10-piece-60575.html
I purchased a 2.6MM drill bit. I have a fixed drilling platform. I made a hole.
Then use the M3 tapping head manually. But it takes too much effort. Then I used a 2.7MM drill bit.
be prone to.
But the screws cannot be tightened. I thought for a moment. Still spent an extra $4. Let the seller of the chassis help me with these tasks.
 

Attachments

  • 钻头.jpg
    钻头.jpg
    74 KB · Views: 176
Compared to QUAD405, QUAD707 has a very significant change. Mainly for improvement.
Measure QUAD707 SR=25V/US using an oscilloscope.
QUAD405 SR=4V/US. Although it does not represent good or bad. But it is obvious that the changes in QUAD707 are significant.
Additionally, QUAD707 is high-frequency. Low frequency. The attenuation rate is smaller. It retains its beautiful sound while also preserving it.
It has significantly improved performance. It was born around 1995.
 

Attachments

  • QUAD707-5.jpg
    QUAD707-5.jpg
    115.3 KB · Views: 159
  • QUAD707-6.jpg
    QUAD707-6.jpg
    102.7 KB · Views: 151
  • QUAD405.jpg
    QUAD405.jpg
    88.8 KB · Views: 157
I don't know what you mean by 'high-frequency. low frequency', but all Quad amplifiers have had similar frequency response since about 1943.
There is a frequency response test chart for APSYS above. You can take a look. Compare QUA405 with QUAD707. The difference is many times greater. Mainly 20-200HZ,
QUAD707 20HZ -0DB。 20000HZ - 0.4DB
QUAD405 20HZ -1 DB, 20000HZ -0.4DB。
In addition, QUAD 707 SR is 6 times higher than 405. So its high-frequency details are more abundant.
 

Attachments

  • QUAD405.jpg
    QUAD405.jpg
    88.7 KB · Views: 121
  • QUAD707-6.jpg
    QUAD707-6.jpg
    102.5 KB · Views: 122
The lover SR of the 405 is given by the RC low-pass between the op amp and the transistor stages. So, on should compare only the SR of the transistor stages.
This is wrong. QUAD405 regardless of the use of any transistor output. His SR is the same. You can use an oscilloscope for measurement. SR is mainly determined by GAIN without feedback. So it can be concluded that the QUAD 707 GAIN is higher than 405 6 times. This is determined by its GAIN. It will not change due to different transistor models.
 
Compared to me. I feel that the 405-2 is good in all aspects. Listening to music is great.
405 sounds good too. The high-frequency is slightly lower.
707 sounds very low-frequency and heavy. There are even some too many. I don't know which one is better.
Perhaps different people have different preferences. Compared to the QUAD series.
I still prefer the L-series amplifier I designed myself. I really like L20V10. Perhaps it's not the best. But I prefer this effect. His voice is quite magnetic.
 
This is not the most important thing. QUAD707 SR is 6 times higher than 405.
So its speed is very fast. It sounds like the dynamic difference is very obvious.
Additionally, 1DB is already very large. There is a 30% difference.
From a DIY perspective. The cost of QUAD707 is almost the same as that of 405.
707 uses 6 output transistors. Better driving performance.
Of course, the 707 may be wider. I can do more miniaturized QUAD707 or QUAD606 in the future, which can replace QUAD405。
 

Attachments

  • IMG_7062.JPG
    IMG_7062.JPG
    240.2 KB · Views: 98
  • IMG_7060.JPG
    IMG_7060.JPG
    263.5 KB · Views: 85
  • Like
Reactions: 1 user
....
I still prefer the L-series amplifier I designed myself. I really like L20V10. Perhaps it's not the best. But I prefer this effect. His voice is quite magnetic.
Hi, I confirm too, the L20V10 is to my ears too, a good amplifier, a true successful design. Best in this L series compared to the Mosfet L series to my ears.
Perhaps miss more heat, higher temperature in some transistors (where ?), better dc polarization, operation point to keep the transistors in a more stable temperature to deliver a more neutral sound ?
(Note my L20V10 with Hypex SMPS - 1.2KW 2x46V)
Keep working ;)
 
Hi, I confirm too, the L20V10 is to my ears too, a good amplifier, a true successful design. Best in this L series compared to the Mosfet L series to my ears.
Perhaps miss more heat, higher temperature in some transistors (where ?), better dc polarization, operation point to keep the transistors in a more stable temperature to deliver a more neutral sound ?
(Note my L20V10 with Hypex SMPS - 1.2KW 2x46V)
Keep working ;)
Yes, L20V10 is very good.
But it doesn't have much history.
I have assembled one QUAD405. But I don't really like it.
I prefer QUAD707. In use. Perhaps its effect will not exceed L20.
But I really like its history. I knew about this series many years ago.
Additionally, I recommend using a circular transformer for L20V10.
I don't really like SMPS.
SMPS has many high-frequency interferences. It will make the transistor very hot.
Of course. SMPS has a significant difference between them. Some are very good. Some are very poor.
Actually, for household use. We only need a low-power one. A circular transformer is sufficient.
For example, AC24-0-24200VA. You can give it a try.
There is actually no difference in the power of transformers. I even used a 50VA transformer to make an amplifier.
Drive speakers exceeding $400000. The same is great.
I think those power data. It's just a promotional advertisement. It's not very useful.
Especially SMPS. In fact, we cannot verify the power of SMPS.
In China. Most 50VA SMPS. The nominal power is 800VA. Sales have exceeded ten years. o_O
And no one noticed that its power was fake.
So what does this mean.
I think the biggest problem is. Actually, 50VA is sufficient for use.
 
No it isn't. It is a factor of 0.89 down or 1.12 up. Do the sums. It is barely audible to an expert listener. 30% is 2.28.

Your slew rate calculation is therefore suspect as well.
SR is not calculated using. Many people here should understand.
It can be seen with the eyes. Because the oscilloscope has already displayed SR 25V/US

If you don't quite understand.
Actually, you can search for the official document data of QUAD707.
Of course, there is also data on QUAD405.
This amplifier was not designed by me. Its data is written by QUAD company.
I'm just measuring. And verify it.
The results are indeed the same as the nominal data of QUAD company. So QUAD's data was not falsified.
QUAD405 SR=4V/US
QUAD707 SR=25V/US
 

Attachments

  • QUAD707-5.jpg
    QUAD707-5.jpg
    115.3 KB · Views: 95