Marantz 250 Repair Testing

[Stoddrob]

I have been repairing a Marantz 250. I made some previous posts on it. One dealt specially with transistors I tested and used. I hope it might be useful to someone.

I will mention again that I originally bought a customized Marantz 250 rebuild kit on eBay from a fellow AK member. The benefit of this “customized” rebuild kit (I didn’t want the filter capacitors) he promoted was the “matched transistors.” This is especially important for the output transistors as well as for the differential pair of Q501/Q502 which are 2N5961 transistors. I had purchased from this Seller before and he seems like a very knowledgeable guy I could trust.

Another reason I purchased the “Complete PLUS ULTRA Rebuild kit” from him was that I previously purchased a full set of matched output transistors from him (last year). This year I purchased a transistor tester and pulled all the output transistors in the Marantz 250 before I started working on it. None of the “matched” transistors were actually matched that close. He wrote numbers on them but the results i got were completely different.

So my plan when I purchased the “custom rebuild kit” was to get another set of output transistors and check and see if they were matched.

The kit arrived and I tested the output transistors. The results I got were nothing like what he had written on them. At least on the NPN transistors. The PNP ones were better. However, since I now had sixteen transistors, I was able to create one full matched set out of two.

He said he uses a Peak Atlas DCA55 tester.

Yesterday, I sensed an issue with the matched differential pairs he provided me in the kit. Since installing them, I had purchased a supply of Fairchild NOS 2N5961 transistors. These are the same ones Marantz originally used. I tested them and created four or five matched sets. I went to my supply and tested them again. I got the exact same results. Further, the hFE gain on most of mine were >500. I chose to reinstall a set of 585 hFE in one channel and a set of 625 hFE in the other. The transistors are also paired much closer (almost perfectly paired) than the “matched” ones from the kit.

I retested the “matched set” transistors I got from the kit Seller on eBay. They were not that closely matched at all. Further, they were in the low 200 hFE. So I’m glad I did that! I shouldn’t have relied on the sellers “expertise.”

Moral of this istory is: Be wary of kits or sellers promising matched components. Do the homework. Find and test the best possible components yourself. It is possible that his Peak Atlas DCA 55 tester was giving him those results. But it seems strange to me that the PNPs were closer to being matched than the NPNs.

I should also mention that I removed, cleaned and tested the BIAS and DC Offset potentiometers in both amp boards. They were in very good condition and the cleaning helped make them much easier to turn. Testing results gives me peace of mind.

DC Offset on both channels is near zero. There is a little jumping around from -1.5mv to +1.5mv but both channels mostly stay under 1mv and will hold at zero for brief periods.

Having said that, I did initial start-up tests with a DBT. I checked all the voltages according to page 8 (10 of 16) from the Marantz 2nd Addendum for the 250. It is specific to the amp board version in my amplifier. I had only one anomaly. It is the -1.4v bridge wire (not sure what else you call it). See the first image below. I got a reading of 12.7v in the left channel and 6.6 in the right channel. Everything else tested basically the same. The pencil notations of test results are not mine.

I decided to move to running putting a signal through the left channel to start and see what kind of sine wave I got using my handheld Oscilloscope/MM/DDS. Note: I am just learning how to use an oscilloscope and decided to experiment using this handheld unit before connecting my Siglent 1202x-e which I have yet to use.

I set the Signal Generator for a Sine Wave at 1k Hz and .1 volts (100mv). I then increased to .2v (200mv). I connected the Signal Generator to the amplifier left channel RCA input. I connected the oscilloscope probe to the left speaker positive output and the probe ground to the speaker negative output. The images below are of the handheld sine wave. I don’t know….but it looks OK to me. The first is at 100mv and second image is at 200mv.

Any comments on the initial results? It doesn’t appear to me that I have major problem after a complete rebuild. Nearly every component on these amp boards has been replaced.

Thanks for looking. Comments are appreciated. I will test the right channel next and then hook the amplifier up to the Siglent.

 

[Jose Jimenez]

Sounds like you are well under control. That is great.

For those diff input transistors, I made bridge circuit to switch between a standard and another device to see if the gain was higher, lower or about equal. That was fun. I really count on the results using the curve tracer. I can switch between two devices and compare accurately. If the circuit board permits, I band or glue the transistors together.

What were the output devices and date code the eBay seller supply to you???!!! I most recently picked up MJ15003/4, MJ15024/5 and MJ21193/4 transistors from Mouser and Newark and found that I liked the MJ15024/5 and MJ21193/4 because between the NPN and PNP I was able to find beta matches easily. Not too long ago I destroyed semiconductors from First Source and that New Jersey company on eBay and was refunded because I brought up the issue on their ISO 9001 certification with the crap they semi me being counterfeit.

 

[Stoddrob]

Jose, I don’t know the date codes but they were Onsemi MJ21195G and MJ21196G. They were good transistors —just not matched. At least not close enough to make me happy.

 

[Stoddrob]

Jose, I don’t know the date codes but they were Onsemi MJ21195G and MJ21196G. They were good transistors —just not matched. At least not close enough to make me happy.

Date Code BM0632 and BM0635

 

[Dave1027]

Peak Atlas vs FNIRSI:

 

[Stoddrob]

Peak Atlas vs FNIRSI:

No doubt Peak Atlas (especially the DCA 75 Pro) is superior to the FNIRSI. I discovered that in short order. Especially when it didn’t identify the transistor pinout. However, in some cases it exactly matched what the Seller had indicated was the gain of the transistor, and on other transistors it was different. But I did not believe the transistors were matched perfectly as the Seller had indicated. I tested other transistors weeks later and I got the exact same results with the FNIRSI. So while it may give different results — at least it’s consistent. Its good to check the health of a transistor. I went by the measurements it gave me in matching transistors. While you might get different test results with a Peak Atlas unit over a FNIRSI, they still should be consistent in terms of being the same (comparable, similar, matched) between a test group. That’s my opinion. I plan on buying a Peak Atlas DCA 75 Pro.

 

[Jose Jimenez]

The Peak Atlas DCA 75 is a good transistor tester for small signal transistors, but it is a toy for accurately testing power transistors. The 2.5mA it uses for collector current to determine beta, heck most amplifiers are biased with more current than that. Unless you have a curve tracer and can put the device under load, may I suggest that you focus on obtaining a piece of test equipment that can actually do the job. I have a few in mind that would be more than adequate to do all your transistor testing. One in particular supplies base current with 50uA (Lo), 10mA (Hi) in pulses in Hi mode to determine beta (gain) for larger devices. Now that's a good start!