Are you using a 360K or 1.2M floppy?

Hmm the specs say "A single 1.2 MB 5.25 inch floppy disk drive was included in the base configuration. "
Added: The original 3000 had Tandy either DOS 3.2 or DOS 3.3 on 360k, you can find images at win world and they include the Tandy "setup.com"...

Hi All, thanks for the replies. It would seem that we are all headed in the same direction. I do note that each time it tells me to "Strike F1 to retry the boot", it accesses the floppy drive, although it will never boot the very bootable floppy in it. This is however a 3 1/2 floppy. Somewhere in the life of this machine, an owner swapped the original 5 1/4 floppy with a 3 1/2 floppy. My GUESS is that when otherwise un-configured, the BIOS is looking for a 5 1/4 floppy drive, and I agree, probably a 360K diskette in it. SO... I have ordered a 5 1/4 floppy on eBay and will try that once it arrives. Else, I fear that the Tandy 3000 may be a "brick" in the short term.

Thanks as well for the pointer; I will head back to WinWorld and poke around more. I believe that this is where I got the Tandy 3000 setup.exe program. If there was more there, I may have missed it. I will take a more thorough look! Thanks again!

It might work as-is with a 720K floppy (or if you format one that way by putting tape over the hole).

Maybe it is a Tandy 3000NL which was the upgraded 3000. It might boot a 720k disk and there are 720k images of Tandy DOS 3.3 at that same place.
Does it have built in VGA (the latter 3000NL's did) versus an Adapter card ?
One thing iirc: older DOS versions must be written properly to floppy or will not boot. DOS 5 and up do not have that issue.

Hi All, I have more or less come to a complete stand still. Nothing seems to work. I am stuck at early BIOS boot, where the Tandy 3000 keeps telling me to insert a bootable diskette and press F1 to retry the boot. At this point I have tried three different disk controllers, I have replaced the 3.5" floppy the machine arrived with with a 5.25" floppy, removed an extraneous network card (perhaps an IRQ or IO port conflict) and tried every size and shape of diskette: full capacity, 720K, 360K, both 3.5 and 5.25. I even added a hard card with some BIOS on it, hoping to work around the issue and when that didn't work, tried connecting an alternate IDE HDD... nothing has caused the machine to deviate from telling me to insert a bootable diskette and press F1. It is very frustrating.

What is even more frustrating is that I HAVE the Tandy 3000 setup utility needed to configure the BIOS to get past this problem, but I need to be running DOS to use it. Of courser, to run DOS I have to have previously run the setup utility. It is a total catch-22 and as a result, the Tandy 3000 is more or less a "brick" at the moment.

Does anyone have any ideas that might lead to a breakthrough? Perhaps a later version of Phoenix BIOS that is available as 27C128-15 ROM chips? Alternately, is there perhaps some minimal debugging monitor/interface built into Phoenix BIOS 1.03 that I could get into and manually load the right values into the battery backed up CMOS parameter space? Finally, I have read of some alternate "generic" BIOS' that I could replace the current BIOS chips with, but I have not been able to track any down.

Just an FYI note, in case it fires off anyone's long lost memories... when I start up the machine, it emits one long and one short beep, and thereafter, each time I press F1, it emits two short beeps. Of course, when I look up Phoenix BIOS beep codes, neither of these combinations is listed! 🙁. It is trying to tell me something; I just don't know what.

I am at a total standstill with this Tandy 3000. Any and all suggestions MUCH appreciated!

Oh, and by the way, it is a Tandy 3000, not an HL or an NL.

Hmm IBM also used Phoenix bios and one long - one short was a board issue back in the XT early AT days....not much help.
Is there a cmos reset jumper ? If CMOS was returned to original state the 1.2Mb floppy should be default IF the bios was properly written... just a thought.
You said seller replaced battery, which should reset the cmos but maybe not fully....

Thanks @horun. I do have the user manual for the Tandy 3000, and all it has to say on this topic is that if the computer beeps when you start it, it usually means that the BIOS settings have been lost due to battery exhaustion. It then recommends putting the Tandy 3000 Utilities diskette (I cannot find such a thing) into the floppy drive and restarting. This IMPLIES that even when reset, the BIOS will recognize and boot the floppy, which I agree is what SHOULD happen with a well designed BIOS, but it is not in my case. The system accesses the floppy but always prompts me to insert a "system diskette" and press F1. I will check for a CMOS reset jumper... thanks for the tip.

I did get very creative today and found an Adaptec SCSI card with its own BIOS extension, thinking maybe I could boot off a SCSI device. No joy. The card was recognized and its BIOS popped up a SCSI utility, but it would not "see" any devices I attached to it. I am not sure WHAT caused that. I moved a known working external Jaz drive onto it (working fine right now on another system I have) but it was not recognized on this system. The Adaptec card declared a timeout on "SCSI Inquiry" or some such and that was it.

This particular card also had a floppy interface, so I wired up the floppy to it. No joy there either. Same behavior. The BIOS accessed the floppy but always prompted for a "system diskette". The BIOS is either faulty, or it is looking for something very specific that is only to be found on a Tandy 3000 Utilities diskette.

The effort continues, but I am losing hope of ever resurrecting this machine. The knowledge needed to do so may have faded out of existing memory. It IS a 35+ year old machine...

What if you just try to boot Disk 1 of MS-DOS 5.0, either in 720K or 360K format, having created a disk directly from the disk image without any modifications?

One way or another it can be fixed. If all else fails, as you suggested using a card with its own BIOS extension, you could put the debug monitor you speak of into the BIOS ROM of such a card.

The saga continues. I purchased an ISA IDE-SD adapter card with the XT-IDE BIOS on board, thinking that I could get control of the system that way, and hopefully boot it off of the SD card. Regrettably, the "chicken and egg" scenario continued. I slotted that card into two different fully operational MS-DOS PCs eventually, each time assuming that I could FDISK and SYS the SD card and then move the booting result to the Tandy 3000.

This effort never got off its feet. No matter what system I put the card into, FDISK would either fail to see it, or would see it, partition it, force a reset and then after the reset, the card would be blank again! I got similar behavior from two different systems! Thinking that maybe the adapter card was bad, I tried viewing the SD card from an old USB SD card reader that I have, connected to a Win98SE system. It too would see the SD card in the reader, but refused to FDISK or format it. No matter how I approached this, the result was always the same... nothing. I finally gave up again, and this time went on the hunt for another ISA card of this type, but sold with an already bootable SD card. I found one that uses CF cards, and comes with a DOS 6.22 bootable CF card. I have ordered that and will have another go at this when it arrives. I am nothing if not persistent!

BTW, for anyone else out there who may be thinking of using XT-IDE BIOS, be aware that it too does not have a built in SETUP utility. I found this out the hard way! I just ASSUMED that a new and modern BIOS would come with built in SETUP. I was severely disappointed. It does not. There is an external configuration program, XTIDECFG, but you need to get the one that matches your version of XT-IDE, since it is highly configurable when being built. I found a generic XTIDECFG, but it was (much) less than useful. So, be aware of this if you are thinking of using XT-IDE BIOS.

If all else fails, I have found an ISA card with two EEPROMs on it that I can program to appear to be option ROMs to the BIOS. The Tandy 3000 BIOS, for all its faults, does recognize option ROMS and passes control to them as it should. So, armed with that card and an EEPROM flasher (haven't looked into THAT yet) I can burn a debug monitor onto an option ROM and manually write into the CMOS memory! Sheesh! That is a lot of work just to get a pseudo SETUP running. However, early in my career, I wrote a very powerful ROM BIOS debug monitor for the 8086 (MON86) and I kept the listings. I still have them after all these years. It is written in Intel's PL/M-86, so I will have to convert it to C, but I am up to that challenge. I hope it does not come to that however!

I'll report back again when the ISA IDE-CF adapter card with bootable DOS 6.22 arrives and I have had a chance to test it.

This is a very early AT clone, battery may only be for RTC, there are two banks of DIP switches onboard which may be for drive configuration. No details of those are given at "the usual suspects" RetroWeb, Stason etc.

Finally, some good news. I purchased one of these on eBay - see the attached image:

https://www.ebay.com/itm/124122830301?hash=it … %3ABFBMlq_ji-lh

and it turned out to be the breakthrough that I had been hoping for... the key that opened the box, so to speak.

I inserted this card into a spare 8-bit ISA slot and powered up. Boom! It booted the CF card immediately and went right to DOS! After all this struggle, I couldn't believe how easy that turned out to be! Of course, now I had DOS running, but no way into or out of the machine... no way to get the Tandy 3000 setup program onto it, but at least I finally had control of the box. I took what seemed to me to be an enormous risk and plugged the CF card into my Mac via a USB CF card reader. I copied the Tandy 3000 setup program, and a few other goodies, onto the card and praying that this hadn't trashed the format of the card, re-inserted it into the machine. Mercifully, it came right back up again and I was able to ...finally... run the Tandy 3000 setup program and configure the BIOS.

Now as far as the floppy goes, this didn't help at all. I configured the BIOS for a 1.44MB 3.5" floppy, which is what the machine has, but the BIOS just seems to ignore that input, and continues to expect a 360K 5.25" floppy as the A:\ drive. To this point, I still don't have a working floppy on the machine. I continue to work on this, but I need a wholly BIOS independent installable floppy disk driver before I will achieve success there... the BIOS isn't going to budge on this - it simply needs to be overridden.

Quite aside from the floppy however, I did manage to configure the hard drive parameters in the BIOS and to my joy, the machine immediately booted from its hard drive on the next restart. Better yet, the CF card now shows up as the D:\ drive! The card itself is removable and booting no longer depends on it, so it has become sort of a "super floppy" - removable storage that I can transfer files back and forth between my Mac and the Tandy 3000 on.

A few other notes. I had an old ATI VGA Wonder ISA video card lying around, and as an experiment, I replaced the original full form factor video card with the VGA Wonder. The result... silence at start up! No more loud long beep then short beep. These beeps were not to be found in the Phoenix BIOS beep code lexicon because they were not coming from the main BIOS! ...they were from the video BIOS on the original video card! That explained a lot.

The VGA Wonder is a multi-technology card, supporting CGA, EGA and VGA, and I have all the software for it. So I will configure it for EGA operation and finally achieve the goal that has been motivating all of this work: an 80286-based EGA test bed for my VE text editor.

Thanks for all your thoughts and inputs - it has all been a very helpful part of this journey to ultimate success.