• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

SAE 2200 Rehab Project - questions

sabrefencer

sabrefencer

AK Subscriber
Subscriber
I have been the owner of an SAE 2200 power amp since 1977. I am preparing to do a full refrurb on this amp. This is an early version (SN below 1000) of the 2200 which had some issues addressed by multiple service bulletins (which I have), and which were fixed on later units by revision of the design. The amp is functioning, with no noticeable audible issues. It was last serviced in the early ‘80s, and this service did not fully/correctly apply the service bulletins. Specifically, the Fairchild output transistors were only changed on the one channel that let out magic smoke in 1981, but the others are still in there and ticking toward their likely demise.

To get a baseline for my work, I did get some “before” measurements of the unit. Visible clipping on the scope was at 30 volts RMS or about 112 watts for both channels driven into 8ohm dummy loads with a 1kHz signal. Distortion at 1kHz measured about 0.2% on both channels at 50 watts. Given the limitations of my HP334A analyzer, these measurements seem fine for a unit that has not seen a tech in 40+ years. They also show that despite the half-way repair, both channels are performing about the same.

DC offset is about 54 mV on both channels, which I think is right at the spec limit.

I measured the bias at the 62 ohm resistors on the driver board, with a 20kHz input signal at 2.8v output to 8ohm dummy loads (1 watt) per SAE service bulletin 2200-01. However, turning the bias pots makes only very small adjustments over a range from about 0.4 vDC to 0.76 vDC, but I can’t get it up near 1v, on either channel. My understanding is that 1.0vDC is the target setting. I measured the resistance of the trim pot that I could get to without removing the output module, and its range is from 1.5 ohms to about 418 ohms, which seems about right for a 500 ohm pot. The 62 ohm resistors read 60.7 and 62.5 ohms (in circuit), which also seems "close enough". Turning the pots also barely moves the needle on the distortion analyzer. Am I measuring this right? Any thoughts on why I can’t get a better range of adjustment?

Do any of the resident SAE 2X00 series gurus such as @llwhtt or @hamrules have any ideas whats going on here?

Gary
 
Register to hide this ad
That DC Offset can be lowered if you want by closer matched input differential pairs and/or making sure those high ohm resistors are close matched, the 15k & 150k (180k ?) ones I believe (Looks to be different values depending when built). I assume you have the BC series transistors in those positions? If so real easy to replace and get that down a lot lower.
 
Last edited:
I don't have that service bulletin so I can comment on what it might be saying or not saying.
I think you want to get the Quiescent Current set first before the bias, the one you have to use what they call "select" resistosr to set.
 
I'm probably not measuring it right. So I'll humbly ask what do I need to be measuring and how? Quiescent current - I assume this is with the amp at idle with no input. Where and how do I measure this? My amp does have "select" resistors on the driver board, they are 750k ohm. Regarding measuring the bias, I have attached the service bulletin referenced in my OP. What is the recommended procedure for measuring bias? In both cases, just to rule out "operator error" exactly where should I be putting the leads and what mode should my meter be in?
 

Attachments

sabrefencer said:
I'm probably not measuring it right. So I'll humbly ask what do I need to be measuring and how? Quiescent current - I assume this is with the amp at idle with no input. Where and how do I measure this? My amp does have "select" resistors on the driver board, they are 750k ohm. Regarding measuring the bias, I have attached the service bulletin referenced in my OP. What is the recommended procedure for measuring bias? In both cases, just to rule out "operator error" exactly where should I be putting the leads and what mode should my meter be in?
Click to expand...
You should have the 68ohm resistors (R35/R37 channel A and R36/R38 channel B), measure across one of them for each channel. What do you come up with, around 1.0vdc?
Far as bias I do not have a distortion analyzer, so I am unable to test the way they call out. @llwht does have this equipment so he could help with doing it that way. For myself I'd check the 62ohm resistor (R59 channel A and R60 channel B) by measuring across it for about 1.1/1.2vdc.
 
Last edited:
I figured it was user error. I was measuring from one leg of the 62ohm resistor to chassis ground. When I measured across (both legs) of the resistors the measurements made more sense. Quiescent and driven (per loads above) dc voltage across the resistors was about 1 volt, and I was able to set both channels to match each other at 1.1.vDC. "Problem" caused by the operator and solved by you. Thank you.
 
When I measure across the 68 ohm resistors I get about 0.42vDC on one channel and 0.75vDC on the other. What is the method to "adjust" this? Is this the function of the "select" resistors?
 
sabrefencer said:
When I measure across the 68 ohm resistors I get about 0.42vDC on one channel and 0.75vDC on the other. What is the method to "adjust" this? Is this the function of the "select" resistors?
Click to expand...
Yes. The higher the select resistors the more voltage you will see. I know of no calculation to figure what is needed. Of late I have been retrofitting a trimmer into the position, like a 2.5 megaohm, to make it easy to adjust. There is also a decade box, I think it's called, that gives all sorts of resistor combo's so you can figure out what is needed and just order the correct ones.
In the past I tried putting in higher resistors and it's a pain, I found you need to make large increases versus small increases in value.
 
If I could fit the trimmer in, I'd probably use a 2.5 - 3 meg and set it at 1 meg for initial turn on and go from there adjusting it.
 
sabrefencer said:
When I measure across the 68 ohm resistors I get about 0.42vDC on one channel and 0.75vDC on the other. What is the method to "adjust" this? Is this the function of the "select" resistors?
Click to expand...
That .42vdc is a little low there, if they were both .75vdc I might just move on.
 
I will pull one leg of each resistor and measure them out of circuit, to see if perhaps one has drifted. I do have a decade resistor box so I could "dial" them in and the fit the nearest standard resistor value. But I like the idea of the trimpot.. Mouser is out of stock, but Digikey has these: 3296W-1-252LF-ND
 
sabrefencer said:
I will pull one leg of each resistor and measure them out of circuit, to see if perhaps one has drifted. I do have a decade resistor box so I could "dial" them in and the fit the nearest standard resistor value. But I like the idea of the trimpot.. Mouser is out of stock, but Digikey has these: 3296W-1-252LF-ND
Click to expand...
That shows as a 2.5K ?. I’d go with a 2.5meg.
 
I also have a few questions about the 62ohm bias resistors. I plan on replacing both of them, since one appears to have been replaced and I want them to match. They measure 60.7 and 62.5 ohms (in circuit). I have seen a reference that these are fuseable, but I can’t confirm this and can’t find 62ohm 1/2 W fuseable resistors. I saw the fuseable reference in a 2400 service manual that I have.

I am thinking of using 62ohm 1W, 1% resistors 279-H4P62RFZA, there is a 1/2W H462RBYA which are 0.1% precision which seems like it might be overkill at $2.24 each. But maybe in this application that's appropriate. Typically I wouldn’t hesitate to go from from a 1/2W to 1W resistor, since if they are functioning as a fuse, the higher wattage rating might not be a good idea.

There don't seem to be too many choices in 62 ohm resistors. Is there a go-to for these bias resistors, or will any 1% metal film resistor work fine?
 
I would also like to get your input on transistor selection:

I already have OnSemi MJ21193/94 for the outputs, and MJE15032/33 for the drivers.

For the other TO-220s on the driver board (Q9,10,11,12) I’m looking at KSA940/KSC2073.

For the TO-92s it has been suggested to use MPSA06 and MPSA56. Would these be a good choice to replace all the TO-92’s in the unit? My unit has 2N5210/2N5087 for the 8 limiter transistors, are these the BC transistors you referred to in post #2? Should I use direct replacements for these, and go with the MPSAs for the others on the driver board, and on the output and power supply board?

I think the 2N5210/2N5087 have higher gain than the MPSAs, but I was not able to tell conclusively from the datasheets I read. I do plan on testing the old ones with my Atlas DCA55 meter when I pull them, and compare gain with the new ones, but I want to make sure I order the right ones in the first place.

I do most of my work on tube gear, and I am still a bit hesitant when it comes to selecting replacement transistors. It's just too easy to replace a 12AX7 with a 12AX7, so I'm spoiled.

Any help from you or @llwhtt would be most appreciated,
 
You must log in or register to reply here.
Back
Top Bottom