• Please review our updated Terms and Rules here

Seagate MFM drive model numbers and differences ?

chris_nh

Experienced Member
Joined
Jan 28, 2019
Messages
372
Location
New Hampshire
Hey everyone, I'm curious to learn more about Seagate MFM drive part number variations ... the usual suspects, ST225, etc.

I've seen a random few drives with "MLC" following the model number, ex: "ST-251 MLC-1" What is the meaning of MLC-1?

I've also seen some part numbers like "ST-251 -1". Same curiosity... what does -1 specify?

Of course the "R" versions are RLL rated drives. How about the "AT" drives? Is the logic board any different or are they the same as the regular models?

These are fascinating drives, and I'm finding that many of them still work quite well.

Thanks to anyone who might have insights on the model number variations!
 
RLL drives have a little more band width than non-RLL otherwise the are the same. The - numbers are usually newer revisions.
AT drives have the controller built in and would be connected directly to the data buss.
The main differences to the drives are things like stepping speed(some have auto-stepping), number of head, number of sectors and sometimes rotational speeds.
Dwight
 
My memory is a bit vauge but as I recall:
R at the end indicates and RLL drive, N at the end indicates SCSI, A at the end indicates IDE. I think sometimes the SCSI and IDE would be the same hardware assembly with different logic boards.

I think -1 or MLC-1 at the end indicates a Seagate "refurbished" unit. I seem to recall these would usually have longer track defect lists.
 
Then there are A/X drives which can be connected to either AT (16-bit) or XT (8-bit) IDE ports. One drive that I know of is only 8-bit IDE, which is the ST325X.
 
I think -1 or MLC-1 at the end indicates a Seagate "refurbished" unit. I seem to recall these would usually have longer track defect lists.

No, MLC stands for Manufacturing Line Code - or something like that. The ST-251 for example has the same stepper as the ST-225, but the ST-251-1 (or ST-251 MLC-1) is fitted with a low-impedance stepper that drops average seek times from around 65 to around 20.
 
No, MLC stands for Manufacturing Line Code - or something like that. The ST-251 for example has the same stepper as the ST-225, but the ST-251-1 (or ST-251 MLC-1) is fitted with a low-impedance stepper that drops average seek times from around 65 to around 20.

Yes, that's exactly what I was looking for. Really, I was wondering if the circuit board from an MLC-1 would be the same ... I found a possible donor board for an st251, but it is on an MLC-1 model. Do you know if the circuit board is different to drive the different stepper motor?

I have a few different models of these drives, I understand RLL/mfm drive metrics well, just needed some clarification of the differences like "mlc" and "-1". Thanks for the excellent answer maxtherabbit.
 
Then there are A/X drives which can be connected to either AT (16-bit) or XT (8-bit) IDE ports. One drive that I know of is only 8-bit IDE, which is the ST325X.

Yes, I believe you are correct for the early 3.5" IDE models. I have a working A/X drive in a 386.
 
No, MLC stands for Manufacturing Line Code - or something like that. The ST-251 for example has the same stepper as the ST-225, but the ST-251-1 (or ST-251 MLC-1) is fitted with a low-impedance stepper that drops average seek times from around 65 to around 20.
Hmm, OK, so was there a code for "refurbished" unit?
 
Part of the story:

temp_sea8ncv340bvgt51786fnd1m-vnb2859.png
 
RLL drives have a little more band width than non-RLL otherwise the are the same.
Dwight

Wrong. RLL has not higher bandwith, the bits per inch resolution on a track is exactly the same. ALso the raw data transfer rate betwen the drive and the harddisc controller is the same. RLL is only a more effective encoding. RLL contains less of sync clock signal, but more data signal, that's all. But that needs a (bit) better magnetic surface quality. That's the difference.
 
Wrong. RLL has not higher bandwith, the bits per inch resolution on a track is exactly the same. ALso the raw data transfer rate betwen the drive and the harddisc controller is the same. RLL is only a more effective encoding. RLL contains less of sync clock signal, but more data signal, that's all. But that needs a (bit) better magnetic surface quality. That's the difference.

If we're going to split hairs here about the meaning of 'band width'.... The bits per inch resolution is higher with RLL (2,7), but the number of potential flux transitions per inch is the same as with MFM, aka RLL (1,3).

You could say RLL (2,7) has a higher bit rate than MFM, but the same normalized baud rate.

In the analog domain this could be represented, as stated by Dwight, as a greater 'band width;' while the highest frequency allowed may not increase, band-width is also increased when the lower frequency limit is extended downwards, even if the upper frequency is not increased. Magnetic recording imposes a high-pass filter; change the high-pass lower cutoff frequency down, and the band width increases, even if the low-pass filter at the upper cutoff doesn't change. See https://en.wikipedia.org/wiki/Run-length_limited

The longer allowed run length effectively extends the lower frequency cutoff downwards; even though the upper frequency is still fixed at a certain rate (due to head gap size), the lower frequency limit being moved downwards gives a larger band width. It's very similar to the massive effective band width increase imposed on video tapes by the use of the luminance NBFM technique in conjunction with the interleaved color-under technique; a tape that could normally only record a bandwidth of 10 octaves, using NBFM an effective bandwidth of over 16 octaves can be successfully recorded and played back. The maximum upper frequency doesn't change; but the lowest effective frequency is lowered dramatically, well below 60Hz.

NBFM, like RLL, is an encoding technique. It does require a good linear 6dB per octave rolloff on the low side, and thus better tape is required, much like better disk media is required to extend the effective bandwidth downwards using RLL encodings (use of both 2,7 and 1,7 is out there).
 
Last edited:
I saw another one today with "PR" following the model number sticker, same style as the "MLC" sticker. Interesting
 
Wrong. RLL has not higher bandwith, the bits per inch resolution on a track is exactly the same. ALso the raw data transfer rate betwen the drive and the harddisc controller is the same. RLL is only a more effective encoding. RLL contains less of sync clock signal, but more data signal, that's all. But that needs a (bit) better magnetic surface quality. That's the difference.

I believe that is what I said.
Dwight
 
I believe that is what I said.
Dwight

I should clarity. Smaller domains on the surface, improve bandwidth. It is not that the data frequency is changed but the timing accuracy of the edges are are improved. This is the same as improved bandwidth.
This is just the same as on a cassette tape. Smaller domains mean it can have higher frequency information. Even if the frequency of the written data is the same it will have less jitter. This amounts to improve bandwidth.
Dwight
 
I found the following interesting explanation of "PR" model code in the ST11R/ST11M manual:

---
Paired Program drives are shipped with a Seagate controller. They
are recognized by a "PR" label on the drive top cover. Paired
drives are shipped formatted with a small DOS partition
containing the installation program, INSTALLR, which will low-
level format and partition your drive. You must supply the DOS
program. The ST11M/ST11R can recognize whether a drive has the
Paired Program software loaded, or has been previously installed.

Paired Program drives with formatted capacities greater than 32
MBytes have Disk Manager partitioning software loaded. This
utility has been customized by Ontrack for Seagate. Refer to
Section 2.1 for special instructions on formatting Paired Program
drives.
---
 
One wonders about the "pairing". Might it be to automatically deal with known bad tracks? Most MFM controllers can easily reformat and partition disk drives. I guess for some people it is kind of intimidating. Getting such things already done and only requiring the OS install might be as much as the average person expected to deal with. Also, for people that will build 10 or more system a day for businesses, it might make sense. In such an environment, time is not free. Something I might fiddle with for a few hours is not something that makes sense in an IT environment. Getting it done quicker might make sense, even for someone that knows what they are doing.
Dwight
 
Hi,

I am not sure what you are specifically looking for. However, if you look at here:

ftp://ftp.seagate.com/pub/techsuppt/

you will find documentation for all seagate drives, incl. their specs. This might help you find out the differences you are looking for.

Best Regards

Jens
 
Back
Top